Mechanisms of Erosion in Rheumatoid Arthritis
CHRISTOPHER T. RITCHLIN, MD,
Clinical Director, Allergy, Immunology and Rheumatology Unit,
University of Rochester School of Medicine and Dentistry,
Rochester, New York, USA
Address reprint requests to Dr. C.T. Ritchlin, University of Rochester School of Medicine and Dentistry, 601 Elmwood Avenue,
Room G-6427J, Rochester, NY 14642-0001.
Irreversible joint destruction characterized by cartilage degradation and bone resorption is the major longterm consequence of inflammatory events initiated early in the course of rheumatoid synovitis. Although the relationship between early symptoms and joint destruction is still not completely understood, a significant amount of osteolysis occurs early in the disease and correlates with disease activity1. This osteolysis manifests in several forms, including localized or juxtaarticular osteopenia and generalized bone loss in the appendicular and axial skeleton1-3. Localized and generalized bone resorption is initiated and maintained by complex pathways. Activation events initiated by cell–cell contact between infiltrating T lymphocytes and monocytes, along with resident synovial lining cells, trigger intercellular signaling cascades2. Elucidation of the molecules involved in these signaling processes has translated into improved therapeutic modalities.
The purpose of this editorial is to describe the molecular and cellular features of rheumatoid cartilage and bone, provide an overview of the molecular basis of bone erosion, and review the clinical efficacy of the new biologic response modifiers (BRM) approved for the treatment of rheumatoid arthritis (RA).
INFLAMMATORY EVENTS IN RHEUMATOID SYNOVIUM
The normal synovial membrane is one to 2 cell layers thick and composed of synoviocytes derived from monocytoid (type A) and mesenchymal (type B) lineages. In the rheumatoid joint, this delicate tissue is transformed into a proliferating cell mass that destroys surrounding tissue and bone4. This remarkable degree of synovial lining hyperplasia is correlated with the influx of infiltrating CD4+ lymphocytes and their subsequent binding to activated endothelial cells, although details of this interaction and the initiating immunologic stimuli are not well understood5. The formation of pannus results from migration of these inflammatory cells into the synovium and neovascularization of synovial tissue. Hyperplasia of synovial cells also plays a critical role in pannus formation6. Indeed, hyperplasia of type B synoviocytes is a major reason contributing to joint inflammation and matrix degradation, which results in the release of effector molecules that attract leukocytes into the joint and which enhances the production of matrix metalloproteinases (MMP)7. Additionally, fibroblasts isolated from rheumatoid joints have been shown to release soluble factors known as cytokines that stimulate fibroblast proliferation, a property not observed in fibroblasts isolated from osteoarthritic joints8.
Cytokines are small proteins that are involved in many aspects of inflammation, including the initiation and amplification of leukocyte recruitment into the joint space. A complex cascade of cytokine mediated events alters the phenotype of the type A and type B synoviocytes, resulting in widespread tissue destruction9. Two main cytokines at the apex of the cytokine network in the rheumatoid joint are tumor necrosis factor (TNF) and interleukin 1 (IL-1). TNF-a is a proinflammatory cytokine produced by monocytes, T lymphocytes, macrophages, and fibroblasts that stimulates prostaglandin E2 (PGE2) and collagenase release from human synovial cells10 and potentiates osteoclast differentiation (osteoclastogenesis) and activation11. IL-1 is typically synthesized and released simultaneously with TNF, and the 2 cytokines share similar biologic properties. Although their functional roles are not entirely coupled, TNF and IL-1 do display highly synergistic and overlapping biological effects. The complex nature of this synergy is not well understood. Moreover, IL-1 and TNF exert autocrine effects on macrophages, thereby leading to upregulated synthesis of both cytokines. IL-1 and TNF bind to distinct receptors and are regulated independently of one another12. However, IL-1 can induce the release of TNF in peripheral blood mononuclear cells in vitro and TNF-like activity in the sera of rabbits13. Additional evidence suggests that TNF is the primary inducer of IL-1 in human synovial cells14, underscoring the pivotal role of TNF as the initiator of downstream proinflammatory pathways15.
There is also evidence that TNF can induce the synthesis of IL-6 by monocytes, T lymphocytes, and fibroblasts. IL-6 shares many biologic properties with IL-1 and TNF; however, there are several important differences. First, IL-6 does not stimulate production of PGE2 and collagenase from human fibroblast and synovial cells12. Second, IL-6 enhances rather than inhibits the production of tissue inhibitor of metalloproteinases (TIMP), the natural antagonist of MMP16. Third, in contrast to the actions of IL-1 and TNF, intraarticular injection of IL-6 has been shown to enhance proteoglycan synthesis, suggesting a protective role for IL-6 in the joint17. The contribution of IL-6 to MMP actions is complex. For example, Ito, et al showed that IL-6 stimulates the production of TIMP, while enhancing IL-1 induced production of metalloproteinase proenzymes18. Thus, while IL-6 participates in the regulation of cartilage homeostasis, the net effect of IL-6 on cartilage in the inflamed joint remains to be determined.
Elevated concentrations of proinflammatory cytokines have been identified in the peripheral blood and joints of patients with RA6. Plasma levels of IL-1 correlated with disease activity and were significantly higher in patients with RA compared with age matched controls19. Similarly, synovial fluid concentrations of IL-6 were significantly greater in RA than in patients with osteoarthritis20. Further, RA synovial fluid levels of TNF correlated with the amount of bone resorbed in joint tissues21. Animal data have shown that the combined blockade of TNF and IL-1 reduces synovial inflammation and joint destruction and preserves bone density in arthritic rats22. A study using porcine cartilage explants showed increased resorption and inhibition of proteoglycan synthesis after treatment with TNF or IL-1. These effects were additive when the explants were treated with the 2 cytokines simultaneously23. Studies in murine models of arthritis have demonstrated suppression of proteoglycan synthesis after intraarticular injection of IL-1a, IL-1ß, and high doses of TNF, further supporting a pleiotropic role for cytokines in rheumatoid joint destruction17.
Both IL-1 and TNF, as well as granulocyte/monocyte-colony stimulating factor (GM-CSF), stimulate fibroblast growth. In addition to its promotion of pannus formation via stimulation of fibroblast growth, GM-CSF also enhances the transcription of TNF24. In turn, TNF and IL-1 stimulate GM-CSF secretion by synovial cells; such regulatory interdependence may contribute to the accelerated rate of pannus formation in the advanced stages of RA6. In addition, TNF and IL-1 increase the expression of cellular adhesion molecules, such as intercellular adhesion molecule-1, endothelial leukocyte adhesion molecule, and vascular cell adhesion molecule, which enhance leukocyte migration into the synovium24.
In summary, animal and human studies indicate that both IL-1 and TNF are dominant cytokines that promote the degradation of articular cartilage in the inflamed joint. These 2 cytokines inhibit the osteogenic properties of bone-remodeling cells, accelerate matrix degradation via the activation of proteolytic enzymes, stimulate osteoclastogenesis, and indirectly promote pannus formation and inflammation via interactions with other cytokines.
A MOLECULAR BASIS FOR THE BREAKDOWN OF CARTILAGE
Because RA has devastating effects on cartilage as the disease progresses, it is worth reviewing some of the characteristics of cartilage. The basic function of cartilage in the joint is to provide a tough, elastic shock absorber that limits friction and resists wear on the bones as the joints move25. Matrix is composed of chondroitin-rich proteoglycan and collagen. Chondrocytes in lacunae synthesize and enzymatically digest matrix, and these are processes that fall out of equilibrium in diseases that destroy articular cartilage. Activation of catabolic enzymes and decreased production of enzyme inhibitors can lead to accelerated matrix breakdown (Figure 1).
Figure 1. Activation of catabolic enzymes can lead to accelerated matrix breakdown. A. Rheumatoid bone stained with antibody to RANK (red). Osteoclasts are prominent in resorption pits at the bone-pannus junction. RANK-positive cells can be seen radiating from vessels in the synovium. B. Hematoxylin and eosin staining of rheumatoid synovium. Mononuclear cells are infiltrating the subsynovium and the lining is hyperplastic.
Two major processes contribute to the cartilage destruction seen in arthritis: decreased synthetic activity of chondrocytes and increased enzymatic degradation of the matrix26. Evidence suggests that cartilage breakdown is mediated by MMP synthesized by transformed fibroblasts and chondrocytes in the synovium and macrophages at the interface of pannus and cartilage9. Additional evidence implicates cathepsins and mast cell proteinases in the degradation of matrix components and osteolysis of bone7,27. IL-1 stimulates metalloproteinase synthesis in synovial cells and chondrocytes28. These MMP exhibit selectivity for the different components of cartilage. For example, collagen is the primary substrate of MMP-1, while gelatin is the primary substrate of MMP-2. MMP-3 degrades proteoglycans, laminin, and fibronectin. However, cells that produce MMP also produce TIMP, suggesting that the regulation of cartilage turnover is dependent on the balance of matrix degradation by MMP and inhibition of MMP by TIMP16. An additional level of regulation is suggested by studies that show that TIMP production is differentially regulated by various cytokines in a cell-specific manner. For example, in endothelial cells, synoviocytes, and chondrocytes, TNF inhibits while IL-6 stimulates the production of TIMP16. Calcitonin also opposes this degradative activity by inhibiting bone resorption29.
BONE LOSS IN RA
The destruction of diarthrodial joints is a prominent manifestation of RA. A high proportion of RA patients develop lesions detectable by magnetic resonance imaging (MRI) of the wrist early in the disease process (median symptom duration 4 months)30. Bone edema, manifested as increased signals on fat-suppressed T2-weighted MRI images, is almost always associated with synovitis, suggesting that synovial inflammation may be a predecessor to bone damage in RA patients31. Radiologic damage progresses at a fairly constant rate over the first 6 years after onset of RA32. Compston, et al have reported that the reduced bone mass in steroid-naive RA patients results from a negative remodeling balance associated with a reduced rate of bone formation33. Other studies have demonstrated a relationship between increased bone resorption and disease activity in RA patients2. Focal bone erosions are associated with increased morbidity and may correlate with the severity of disease29. In one study, a majority of patients experienced severe morbidity, with lower grip strength and functional capacity in over 90% of patients, and an increase in mortality relative to expected rates for age and sex matched individuals34. Functional disability was evident at baseline and continued to worsen with time34. Further, the pain and functional impairment associated with RA have substantial effects on quality of life and medical costs35,36, as well as on morbidity and mortality.
OSTEOCLASTS AS MEDIATORS OF BONE RESORPTION
Normal remodeling of bone depends on the coordinated activity of bone-forming and bone-resorbing cells. Osteoclasts, multinucleated cells derived from the monocyte/macrophage line (Figure 2), are primarily responsible for bone resorption. Under normal physiologic conditions, mature osteoclasts are solely responsible for bone resorption in matrix lacunae37. Osteoclastic resorption of bone requires the cells to adhere to the bone surface, an interaction enhanced by increased osteoclast number38. The short lifespan (about 2 weeks) of osteoclasts suggests that factors related to the formation and maturation of these cells represent important mechanisms for regulating their activity39.
Figure 2. Osteoclast precursors originate from hematopoietic stem cells and myeloid progenitor cells residing in the bone marrow. The osteoclast precursors arise directly from a subpopulation of circulating CD14+ monocytes that increase in response to tumor necrosis factor (TNF). The CD14+ monocytes can differentiate into: tissue macrophages in response to macrophage-colony stimulating factor (M-CSF), dendritic cells in response to granulocyte/monocyte-colony stimulating factor, and IL-4, or osteoclast precursors in response to TNF and presumably other factors. Upon entering the inflamed joint, osteoclast precursors encounter relatively high expression of RANK ligand (RANKL) and low levels of the inhibitory decoy receptor osteoprotegerin (OPG). RANKL and macrophage colony stimulating factor (M-CSF) promote differentiation of osteoclast precursors into mature osteoclasts that resorb bone. Adapted with permission, from Athanasou. J Bone Joint Surg Am 1996;78:1096-112.
Histologic analyses suggest that, in addition to their role as mediators of normal bone turnover, osteoclasts also mediate focal bone resorption in RA3. Osteoclasts were originally identified by acid phosphatase staining in rheumatoid joints showing erosion at the junction between cartilage and pannus40. Osteoclasts have been identified more specifically by tartrate-resistant acid phosphatase staining and expression of calcitonin receptor mRNA in rheumatoid lesions29,41. Further, osteoclast precursors have been identified in bone resorption lacunae adjacent to invading pannus27,29. Together with data suggesting that other cell types (e.g., macrophages) at the bone–pannus junction have a limited ability to resorb mineralized bone29, these findings suggest that osteoclasts may be the cell type primarily responsible for bone loss in patients with RA.
The differentiation of osteoclast precursors into mature, bone-resorbing cells depends on a network of cytokines that includes TNF, IL-1, and IL-6. These cytokines have been localized to rheumatoid synovium at the cartilage–pannus junction16. Recently, an additional member of the TNF family of cytokines was discovered by 2 independent teams; the factor was named either osteoprotegerin ligand, osteoclast differentiation factor, or receptor activator of nuclear factor kB ligand (RANKL). RANKL is synthesized by osteoblasts and bone stromal cells, and in conjunction with macrophage-colony stimulating factor (M-CSF), is an essential factor for osteoclast differentiation29,42,43. In an animal model, RANKL has been shown to induce osteoclastogenesis and bone loss44. RANKL binds to its cell-surface receptor (RANK) located on dendritic, T lymphocytes, osteoclast precursors, and osteoclasts (Figure 1). In addition, a soluble "decoy" receptor for RANKL has been identified, termed osteoprotegerin45. This "decoy" receptor blocks the binding of RANKL, thereby preventing RANK activation and subsequent osteoclastogenesis (Table 1)29. Together with data showing that blockade of RANKL signaling prevents destruction of bone and cartilage in arthritic rats44, the observation of RANKL transcription in synovial tissue derived from RA patients suggests that this osteoclast differentiation factor may play a key role in promoting bone loss in the rheumatoid joint27.
Additional cytokines such as IL-1a, IL-1ß, and IL-11 may also contribute to the proinflammatory cascade by binding to osteoblasts and stimulating the release of factors that, in turn, lead to the differentiation of osteoclasts29. One of these factors is M-CSF, which has been shown to be uniquely essential in osteoclastic differentiation46.
TRADITIONAL TREATMENT OF RA
The treatment of RA has centered on relieving the symptoms of inflammation, improving overall function, and preventing disease progression. However, there is little evidence supporting the concept that therapies that improve pain and mobility significantly lessen joint destruction and disease progression47. Enthusiasm for the step-up or pyramid approach to treating RA waned when observational and MRI studies revealed that functional decline and joint erosions emerge early in the disease30. This realization led to the use of standard disease-modifying antirheumatic drugs (DMARD) early in the disease48. The use of combination DMARD regimens also became popular48,49. Despite this aggressive strategy, only modest inhibition of joint destruction has been obtained in trials with individual DMARD, such as sulfasalazine, hydroxychloroquine, and cyclosporine50-53. Based on several comparative trials, a consensus has emerged suggesting that treatment with methotrexate (MTX) limits joint damage more effectively than treatment with the other traditional DMARD54. Initiation of treatment with MTX early in the course of RA coupled with rapid escalation to a target dose of 17.5 mg/week significantly reduced joint erosions and progression of the disease55. Longterm treatment with MTX, however, has been associated with significant toxicity and treatment discontinuation over a 5 year period of up to 75% of patients56,57.
Leflunomide inhibits dihydroorotate dehydrogenase, an enzyme involved in de novo pyrimidine synthesis58. The active metabolite of leflunomide, A77 1726, inhibits DNA synthesis in lymphocytes by limiting the pool of available pyrimidines, thus resulting in an antiproliferative effect. Decreases in T cell proliferation may have antiinflammatory effects via a decrease in proinflammatory cytokines such as IL-1 and TNF59. Leflunomide has been shown to be similar in efficacy to sulfasalazine and MTX in reducing swollen/tender joint counts, as well as in decreasing radiographic progression as assessed by erosions and joint space narrowing60,61. Concern has been mounting recently, however, on a possible association between leflunomide and liver toxicity62-65.
THE EMERGENCE OF BIOLOGIC AGENTS IN RA
Advances in the understanding of the pathophysiology of RA have resulted in the development of 3 new agents with proven efficacy in halting radiographic progression of the disease: etanercept, infliximab, and anakinra. These agents have all been approved for the treatment of RA. Etanercept and infliximab are TNF-neutralizing agents (Figure 3). Etanercept is a human fusion protein produced by recombinant DNA technology and consists of 2 copies of the soluble form of the p75 TNF receptor (p75 TNFR) fused to the Fc portion of human IgG166. Etanercept blocks the biologic activity of TNF by competitively inhibiting the association of TNF with its cell-surface receptors, thereby mimicking the action of endogenous soluble TNFR that neutralize free TNF67. Infliximab is a chimeric monoclonal antibody that binds with high affinity and selectivity to TNF68. Like etanercept, infliximab prevents the association of TNF with its cell-surface receptor, thereby blocking the signaling activity of TNF. Unlike etanercept, infliximab may cause complement-mediated lysis of TNF-expressing cells69 and does not neutralize the activity of the related proinflammatory cytokine lymphotoxin-a, which also exerts biologic actions via TNFR70. The clinical significance of this finding is under investigation.
Figure 3. Neutralization of proinflammatory cytokines by etanercept, infliximab, and anakinra. Activation of tumor necrosis factor receptors (TNFR) and interleukin receptors (IL-1R) initiates signaling events that lead to processes responsible for joint erosion. Etanercept is a human fusion protein consisting of 2 p75 TNFR linked to the Fc region of human IgG. Infliximab is a monoclonal antibody consisting of human constant and murine variable regions. Both etanercept and infliximab neutralize activation of TNFR, with the former mimicking endogenous soluble TNFR and the latter functioning as a high affinity antibody, to prevent binding of TNF to its cell-surface receptor. Anakinra mimics the action of the endogenous IL-1R antagonist (IL-1Ra) by binding to the IL-1 cell-surface receptor, thereby preventing its activation by IL-1. Anakinra differs from IL-1Ra by the addition of a single methionine residue at its amino terminus. Target cells include synoviocytes, chondrocytes, osteoblasts, osteoclast precursors, and endothelial cells.
Anakinra is an IL-1 receptor antagonist (IL-1Ra) synthesized by recombinant DNA technology. Anakinra competitively inhibits binding of IL-1 to its cell-surface receptor, thus blocking the biologic activity of IL-171,72. Anakinra closely resembles naturally occurring IL-1Ra with respect to structure and prevents interactions between IL-1 and cells involved in inflammation71,73,74.
Clinical trials have provided convincing evidence for the efficacy of the new biologic agents in retarding progression of RA. Etanercept has been shown to improve the inflammatory symptoms of RA and to attenuate radiographic progression of the disease55,67. Further, patients who received etanercept had more rapid improvement than those receiving MTX55. In a 2-year followup study of 512 patients receiving monotherapy of either etanercept or MTX, 72% of etanercept patients achieved American College of Rheumatology 20% (ACR 20) improvement criteria versus 59% of MTX patients (p = 0.005). Etanercept patients also had better outcomes as measured by Sharp scale total score and erosion score (mean change 1.3 and 0.66 units, respectively, vs 3.2 and 1.86 units; p = 0.001). More etanercept patients had improvement in Health Assessment Questionnaire disability index (55% vs 37%; p < 0.001)75. Maini, et al observed a rapid reduction in disease activity in patients who were treated with infliximab and who had failed to respond adequately to previous treatment with MTX. Although infliximab produced dramatic results, one disadvantage of this agent is that it must be used concomitantly with MTX76. Further, many patients become tolerant to the medication, requiring either dose escalation or more frequent administration. Clinical trials with anakinra have also shown that this agent reduces the rate of joint destruction and improves measures of disease activity. Moreover, patients who received longterm (1 year) treatment with anakinra continued to show improvement through the study endpoint77.
Although overwhelming evidence suggests that the cytokine family represents a rational target for treating RA, the choice of which individual cytokine/receptor system to target for optimal therapeutic effects remains a central issue. Studies to date suggest that TNF-neutralizing agents produce better symptomatic and functional relief than IL-1-neutralizing agents. In both human studies and animal models of arthritis, the effect of anakinra on measures of inflammation was relatively modest77-79. Intermittent dosing with subcutaneous administration coupled with a short half-life (4 to 6 h)71 may have contributed to the modest effects, insofar as maximal effects in animal studies necessitated continuous infusion of anakinra77. In contrast, etanercept has produced more impressive effects on pain, inflammation, and radiographic progression in RA patients80. However, it is not possible to compare the efficacy of anakinra with that of anti-TNF agents, owing to differences in study design, patient population, and outcome measures81,82. Regrettably, combination of anakinra and etanercept in RA patients did not significantly improve clinical efficacy and was associated with a high rate of infections. For example, neutralization of TNF inhibits production of IL-1, IL-6, and GM-CSF in synovial cells derived from RA patients; neutralization of IL-1, however, does not decrease the production of TNF83,84. These experiments suggest that TNF may be the primary inducer of IL-1 and other proinflammatory cytokines, thus representing the crucial mediator in a cascade that leads to diminished synthetic activity by articular chondrocytes and enzymatic destruction of the matrix.
The clinical efficacy and potency of TNF and IL-1-neutralizing biologic reagents, in not only alleviating RA symptoms, but also in retarding the progression of the disease, may have profound implications for other cytokine/receptor systems that potentiate RA synovitis, including interferon-a, interferon-ß, IL-6, and IL-1785. Current studies of RA-derived tissues and in arthritic mouse models have identified other potential therapeutic targets such as IL-15, an IL-2-like cytokine, and IL-18, an IL-1-like cytokine, as important mediators in the proinflammatory response underlying RA pathogenesis86. Evidence from these studies shows elevated concentrations of both cytokines in RA synovial fluids, as well as a synergistic effect of IL-15 and IL-18 to induce TNF expression in vitro. In converse experiments, however, it has been shown that TNF can similarly upregulate IL-15 and IL-18 synthesis, illustrating the complexity in establishing cytokine hierarchy during RA inflammatory responses87. Although the details of this hierarchy remain incomplete, TNF is clearly central to the pathogenesis of inflammatory osteolysis and acts as the primary trigger in the early phases of synovitis.
The erosive processes leading to joint destruction in RA arise from a complex network of cytokines that mediate signaling among lymphocytes, fibroblasts, chondrocytes, and synovial cells. Advances in understanding the role of cytokines in the inflammatory process have led to newer DMARD that have the potential to dramatically improve the treatment of this chronic disease. IL-1Ra primarily affects secretion of metalloproteinases, degradation of proteoglycans, and osteoclast activation, thereby reducing damage to cartilage and bone. Clinically, this manifests as a reduction in radiographic disease progression with only modest improvements in inflammation. Etanercept and infliximab neutralize the effects of TNF, which may play more of a pivotal role in blocking synovial inflammation than IL-1Ra. By virtue of blocking critical upstream activation events in the cytokine cascade, these agents can produce dramatic improvements in both inflammation and radiographic damage.
Further, these agents improve quality-of-life assessments in patients with RA67,88. Additional studies comparing biologic agents as monotherapy and in combination with conventional DMARD and other biologic response modifiers will be necessary to determine the most effective and safe treatment regimen for longterm treatment of rheumatoid synovitis.
10. Dayer JM, Beutler B, Cerami A. Cachectin/tumor necrosis factor stimulates collagenase and prostaglandin E2 production by human synovial cells and dermal fibroblasts. J Exp Med 1985;162:2163-8. [MEDLINE]
11. Bertolini DR, Nedwin GE, Bringman TS, Smith DD, Mundy GR. Stimulation of bone resorption and inhibition of bone formation in vitro by human tumour necrosis factors. Nature 1986;319:516-8. [MEDLINE]
13. Ikejima T, Okusawa S, Ghezzi P, van der Meer JW, Dinarello CA. Interleukin-1 induces tumor necrosis factor (TNF) in human peripheral blood mononuclear cells in vitro and a circulating TNF-like activity in rabbits. J Infect Dis 1990;162:215-23. [MEDLINE]
15. Nawroth PP, Bank I, Handley D, Cassimeris J, Chess L, Stern D. Tumor necrosis factor/cachectin interacts with endothelial cell receptors to induce release of interleukin 1. J Exp Med 1986;163:1363-75. [MEDLINE]
16. Shingu M, Nagai Y, Isayama T, Naono T, Nobunaga M. The effects of cytokines on metalloproteinase inhibitors (TIMP) and collagenase production by human chondrocytes and TIMP production by synovial cells and endothelial cells. Clin Exp Immunol 1993;94:145-9. [MEDLINE]
17. van de Loo FA, Joosten LA, Van Lent PL, Arntz OJ, van den Berg WB. Role of interleukin-1, tumor necrosis factor alpha, and interleukin-6 in cartilage proteoglycan metabolism and destruction. Effect of in situ blocking in murine antigen- and zymosan-induced arthritis. Arthritis Rheum 1995;38:164-72. [MEDLINE]
18. Ito A, Itoh Y, Sasaguri Y, Morimatsu M, Mori Y. Effects of interleukin-6 on the metabolism of connective tissue components in rheumatoid synovial fibroblasts. Arthritis Rheum 1992;35: 1197-201. [MEDLINE]
20. Houssiau FA, Devogelaer J-P, Van Damme J, Nagant de Deuxchaisnes C, Van Snick J. Interleukin-6 in synovial fluid and serum of patients with rheumatoid arthritis and other inflammatory arthritides. Arthritis Rheum 1988;31:784-8. [MEDLINE]
21. Neidel J, Schulze M, Lindschau J. Association between degree of bone-erosion and synovial fluid-levels of tumor necrosis factor alpha in the knee-joints of patients with rheumatoid arthritis. Inflamm Res 1995;44:217-21. [MEDLINE]
22. Feige U, Hu Y-L, Gasser J, Campagnuolo G, Munyakazi L, Bolon B. Anti-interleukin-1 and anti-tumor necrosis factor-alpha synergistically inhibit adjuvant arthritis in Lewis rats. Cell Mol Life Sci 2000;57:1457-70. [MEDLINE]
24. Heidenreich S, Gong J-H, Schmidt A, Nain M, Gemsa D. Macrophage activation by granulocyte/macrophage colony-stimulating factor. Priming for enhanced release of tumor necrosis factor-alpha and prostaglandin E2. J Immunol 1989;143:1198-205. [MEDLINE]
25. Rosenberg AE. Skeletal system and soft tissue tumors. In: Cotran RS, Kumar V, Robbins SL, editors. Pathologic basis of disease. Philadelphia: W.B. Saunders; 1994:1213-71.
26. van den Berg WB. Arguments for interleukin 1 as a target in chronic arthritis. Ann Rheum Dis 2000;59 Suppl 1:i81-4.
30. McQueen FM, Stewart N, Crabbe J, et al. Magnetic resonance imaging of the wrist in early rheumatoid arthritis reveals a high prevalence of erosions at four months after symptom onset. Ann Rheum Dis 1998;57:350-6. [MEDLINE]
31. 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;42:1706-11. [MEDLINE]
32. Hulsmans HM, Jacobs JW, van der Heijde DM, van Albada-Kuipers GA, Schenk Y, Bijlsma JW. The course of radiologic damage during the first six years of rheumatoid arthritis. Arthritis Rheum 2000;43:1927-40. [MEDLINE]
34. Pincus T, Callahan LF, Sale WG, Brooks AL, Payne LE, Vaughn WK. Severe functional declines, work disability, and increased mortality in seventy-five rheumatoid arthritis patients studied over nine years. Arthritis Rheum 1984;27:864-72. [MEDLINE]
41. Gravallese EM, Harada Y, Wang JT, Gorn AH, Thornhill TS, Goldring SR. Identification of cell types responsible for bone resorption in rheumatoid arthritis and juvenile rheumatoid arthritis. Am J Pathol 1998;152:943-51. [MEDLINE]
42. Yasuda H, Shima N, Nakagawa N, et al. Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 1998;95:3597-602. [MEDLINE]
45. Yasuda H, Shima N, Nakagawa N, et al. Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 1998;139:1329-37. [MEDLINE]
46. Pollard JW, Stanley ER. Pleiotropic roles for CSF-1 in development defined by the mouse mutation osteopetrotic. In: Wassarman PM, editor. Advances in developmental biochemistry. New York: JAI Press Inc.; 1996:153-93.
47. Callahan LF, Pincus T, Huston JWI, Brooks RH, Nance EP, Kaye JJ. Measures of activity and damage in rheumatoid arthritis: depiction of changes and prediction of mortality over five years. Arthritis Care Res 1997;10:381-94. [MEDLINE]
48. Hochberg MC. Early aggressive DMARD therapy: the key to slowing disease progression in rheumatoid arthritis. Scand J Rheumatol 1999;112 Suppl:3-7.
51. Forre O. Radiologic evidence of disease modification in rheumatoid arthritis patients treated with cyclosporine. Results of a 48-week multicenter study comparing low-dose cyclosporine with placebo. Norwegian Arthritis Study Group. Arthritis Rheum 1994;37: 1506-12. [MEDLINE]
53. Boers M, Verhoeven AC, Markusse HM, et al. Randomised comparison of combined step-down prednisolone, methotrexate and sulphasalazine with sulphasalazine alone in early rheumatoid arthritis. Lancet 1997;350:309-18. [MEDLINE]
54. Rich E, Moreland LW, Alarcón GS. Paucity of radiographic progression in rheumatoid arthritis treated with methotrexate as the first disease modifying antirheumatic drug. J Rheumatol 1999;26:259-61. [MEDLINE]
55. 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;343:1586-93. Published errata appear in 2001;344:76; and 2001;344:240. [MEDLINE]
56. Furst DE, Erikson N, Clute L, Koehnke R, Burmeister LF, Kohler JA. Adverse experience with methotrexate during 176 weeks of a longterm prospective trial in patients with rheumatoid arthritis. J Rheumatol 1990;17:1628-35. [MEDLINE]
58. ARAVA [package insert]. Kansas City, MO: Aventis Pharmaceuticals Inc.; 2002.
60. Smolen JS, Kalden JR, Scott DL, et al, for the European Leflunomide Study Group. Efficacy and safety of leflunomide compared with placebo and sulphasalazine in active rheumatoid arthritis: a double-blind, randomised, multicentre trial. Lancet 1999;353:259-66. [MEDLINE]
61. Sharp JT, Strand V, Leung H, Hurley F, Loew-Friedrich I, on behalf of the 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;43:495-505. [MEDLINE]
64. Jakez-Ocampo J, Richaud-Patin Y, Simon JA, Llorente L. Weekly dose of leflunomide for the treatment of refractory rheumatoid arthritis: an open pilot comparative study. Joint Bone Spine 2002;69:307-11. [MEDLINE]
65. Osiri M, Shea B, Robinson V, et al. Leflunomide for treating rheumatoid arthritis [Cochrane Review]. Cochrane Database Syst Rev 2003;CD002047.
66. Enbrel® [package insert]. Seattle: Immunex Corporation; 2002.
67. Moreland LW, Baumgartner SW, Schiff MH, et al. Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. N Engl J Med 1997;337:141-7. [MEDLINE]
68. Remicade® [package insert]. Malvern, PA: Centocor, Inc.; 2002.
69. Lorenz H-M, Antoni C, Valerius T, et al. In vivo blockade of TNF-alpha by intravenous infusion of a chimeric monoclonal TNF-alpha antibody in patients with rheumatoid arthritis. Short term cellular and molecular effects. J Immunol 1996;156:1646-53. [MEDLINE]
71. Kineret [package insert]. Thousand Oaks, CA: Amgen Inc.; 2001.
73. Smith RJ, Chin JE, Sam LM, Justen JM. Biologic effects of an interleukin-1 receptor antagonist protein on interleukin-1-stimulated cartilage erosion and chondrocyte responsiveness. Arthritis Rheum 1991;34:78-83. [MEDLINE]
75. Genovese MC, Bathon JM, Martin RW, et al. Etanercept versus methotrexate in patients with early rheumatoid arthritis: two-year radiographic and clinical outcomes. Arthritis Rheum 2002;46: 1443-50. [MEDLINE]
76. Maini R, St. Clair EW, Breedveld F, et al, for the ATTRACT Study Group. 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;354:1932-9. [MEDLINE]
78. Lewthwaite J, Blake SM, Hardingham TE, Warden PJ, Henderson B. The effect of recombinant human interleukin 1 receptor antagonist on the induction phase of antigen induced arthritis in the rabbit. J Rheumatol 1994;21:467-72. [MEDLINE]
79. Cohen S, Hurd E, Cush J, et al. Treatment of rheumatoid arthritis with anakinra, a recombinant human interleukin-1 receptor antagonist, in combination with methotrexate: results of a twenty-four-week, multicenter, randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2002;46:614-24. [MEDLINE]
81. St. Clair EW. Tides of inflammation: impact of biologics. J Rheumatol 2002;29 Suppl 65:22-6.
82. Fleischmann RM. Examining the efficacy of biologic therapy: are there real differences? J Rheumatol 2002; 29 Suppl 65:27-32.
83. Haworth C, Brennan FM, Chantry D, Turner M, Maini RN, Feldmann M. Expression of granulocyte-macrophage colony-stimulating factor in rheumatoid arthritis: regulation by tumor necrosis factor-alpha. Eur J Immunol 1991;21:2575-9. [MEDLINE]
84. Butler DM, Maini RN, Feldmann M, Brennan FM. Modulation of proinflammatory cytokine release in rheumatoid synovial membrane cell cultures. Comparison of monoclonal anti TNF-a antibody with the interleukin-1 receptor antagonist. Eur Cytokine Netw 1995;6:225-30. [MEDLINE]