Rheumatoid arthritis therapy after tumor necrosis factor and interleukin-1 blockade.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 48, No. 12, December 2003, pp 3308–3319 DOI 10.1002/art.11358 © 2003, American College of Rheumatology REVIEW Rheumatoid Arthritis Therapy After Tumor Necrosis Factor and Interleukin-1 Blockade Kurt Redlich,1 Georg Schett,1 Günter Steiner,2 Silvia Hayer,1 Erwin F. Wagner,3 and Josef S. Smolen2 DMARD (2). In this trial, MTX showed a benefit that was not dissimilar to that of the biologic agent, although some subanalyses and extension trials favored the TNF blocker (2,9). However, there are no head-to-head trials comparing each biologic agent separately, and thus we do not know if one of them may be better than another. Importantly, 2 new trials have revealed that MTX ⫹ infliximab is superior to MTX alone in early RA, and equally important, etanercept ⫹ MTX is superior to either MTX or etanercept alone in established RA. These data suggest that combination of a TNF blocker with MTX may be the most efficacious therapeutic approach at present. However, the results of these studies have not yet been published. Limitations. As much as these new targeted therapies have expanded the therapeutic armamentarium and advanced the field of RA therapy, they nevertheless do have, in addition to their potential adverse effects, limited efficacy (10). In fact, only a small proportion of patients achieve 70% improvement according to the American College of Rheumatology (ACR70) clinical response criteria, and remissions are exceedingly rare. Moreover, the frequency of patients who do not achieve even the weakest response, namely an ACR20 (20% improvement) response, amounts to 28–58% (Table 1). Taking into account that determination of the ACR response, by virtue of its composite nature (14), may over- or underestimate treatment effects, and considering that the determination of a European League Against Rheumatism (EULAR) improvement response as a continuous variable may better reflect therapeutic responses (15,16), the frequency of unresponsiveness according to the EULAR criteria is still high (17). However, in this context it should be noted that patients who are entered into clinical trials of biologic agents have already shown nonresponsiveness to other DMARDs alone, mostly in regimens including MTX with one exception (2), and that the earlier DMARD Introduction The introduction of novel disease-modifying antirheumatic drugs (DMARDs), including biologic agents, has set new standards in the treatment of rheumatoid arthritis (RA). These agents not only modify disease (hence the name DMARDs) in terms of influencing its signs and symptoms, achieving high responder rates, and retarding radiographic progression (1–6), but also have made it possible to translate recent pathogenetic insights into clinical practice (7,8). The success and limitations of the novel therapies Efficacy. In clinical trial settings, the new agents have been efficacious, particularly in patients whose disease had failed to show improvement with prior DMARDs. Although many rheumatologists believe that the tumor necrosis factor (TNF) blockers are more efficacious than traditional DMARDs, this is still unproven, since most trials have tested the new compounds by comparing them with placebo alone or placebo ⫹ methotrexate (MTX) or using them in combination with MTX in patients whose disease remains active despite MTX therapy. There exists only one published head-tohead trial of a biologic agent versus a traditional Supported in part by the ICP project of the Austrian Ministry of Education and Research and the City of Vienna, and by the START prize from the Austrian Research Council (Fonds zur Förderung der wissenschaftlichen Forschung). 1 Kurt Redlich, MD, Georg Schett, MD, Silvia Hayer, MD: University of Vienna, Vienna, Austria; 2Günter Steiner, PhD, Josef S. Smolen, MD: University of Vienna and Center of Molecular Medicine, Austrian Academy of Sciences, Vienna, Austria; 3Erwin F. Wagner, PhD: Research Institute of Molecular Pathology, Vienna, Austria. Address correspondence and reprint requests to Josef S. Smolen, MD, Department of Internal Medicine III, Division of Rheumatology, University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria. E-mail: email@example.com. Submitted for publication April 2, 2003; accepted in revised form August 20, 2003. 3308 RA THERAPY AFTER TNF AND IL-1 BLOCKADE 3309 Table 1. Frequency of nonresponders according to the American College of Rheumatology 20% and 50% improvement criteria (ACR20 and ACR50, respectively) in clinical trials with tumor necrosis factor or interleukin-1 blockers Compound Infliximab* Etanercept* Adalimumab* PEG-TNFRI† Anakinra* Etanercept ACR20 nonresponder, % Placebo nonresponder, % ACR50 nonresponder, % Duration of treatment, weeks Reference 48 29 34 50 58 28 83 73 85 74 77 35‡ 67 61 46 80 76 35 54 24 24 12 24 52 4 11 5 12 13 2 * With concomitant methotrexate. † Pegylated recombinant methionyl human soluble tumor necrosis factor type I. ‡ Methotrexate used as comparator. courses appear to be more effective than subsequent ones (18). With regard to retardation of radiographic progression, it is evident that the biologic agents, especially in combination with MTX, may achieve results rarely seen before. Nevertheless, both with etanercept and anakinra, there is a mild slope of progression suggesting that disease continues to progress, albeit at lower levels (2,3). Even when considering the most impressive data published to date on retardation of radiologic progression in RA, namely trials of infliximab (4), there are many patients whose joint destruction still continues. What do these results mean? 1) Do they mean that we do not use the right doses of biologic agents in all patients? This is unlikely, since no significant differences in retardation of radiographic joint damage were seen among several doses of infliximab, with the lowest dose being as efficient as the highest one (⬎6-fold dose difference) (4). 2) Do they mean that more than 1 proinflammatory cytokine is involved in driving the rheumatoid process? This is a possibility, since inhibition of the activities of interleukin-1 (IL-1) also retards joint destruction, and there may be TNF-independent IL-1 activation (19). Furthermore, IL-6 may also be a player in this respect (20). 3) Do they mean that other mechanisms, upstream or downstream of the production of proinflammatory cytokines, may be of importance? This will be one of the focuses of this review. It is evident from the data in the literature that use of TNF blockers combined with MTX may be the most powerful strategy currently available for treating RA. However, the data shown in Table 1 reveal that more needs to be done to attain full control of RA in all patients. We need new therapies in order to achieve better relief from the signs and symptoms of RA than is currently possible, to fully revert RA from a destructive to a nondestructive joint disease, to induce a complete response or remission, which is currently only rarely achieved with the biologic agents, and, ideally, to lead to a cure. To address the issue of potential new therapies, we have divided the therapeutic targets into those upstream of TNF and IL-1, those occurring downstream of the proinflammatory cytokines, and events deemed to be active in parallel with TNF and IL-1. The selection of the targets is, by no means, comprehensive. Nevertheless, the selected targets ought to serve as examples for future approaches. Moreover, in this context, we will also discuss changes in therapeutic strategies. Targets upstream of TNF and IL-1 Role of T cells. There is a continuing debate on the involvement of T cells in the pathogenesis of RA. Several investigators have suggested that the major pathogenetic events in RA involve the macrophage and fibroblast-like cells (21). There is significant supporting evidence for such an assumption, and the major inflammatory events, which include joint destruction, are clearly mediated by these cells. Nevertheless, the major histocompatibility complex (MHC) association with RA (22), the production of significant amounts of T cell– derived cytokines (23) when compared with active immunization with T cell–dependent antigens, and the response to therapies directed or partly directed at T cells, such as cyclosporin A (24) and leflunomide (6), suggest that the involvement of T cells in the pathogenic process could not be easily dismissed. Further support for such involvement stems from data showing the involvement of cell contact between T cells and macrophages as one mechanism for the secretion of proinflammatory cytokines (25). Finally, in situations of nonubiq- 3310 REDLICH ET AL Figure 1. Illustration of rheumatoid arthritis synovial membrane, demonstrating differential localization of stress- or mitogen-activated protein kinases. The upper right section of the sphere shows the different compartments of the synovial membrane. EC ⫽ endothelial cells; PV ⫽ perivascular region; SSL ⫽ synovial sublining layer; SLL ⫽ synovial lining layer. The upper left section and the 4 boxed images show the predominant localizations of p38 MAPK, JNK, and ERK activation. Different cell types of the synovium are illustrated in the lower section of the sphere. uitous and even of ubiquitous autoantigen expression, an arthritogenic or autoimmune response is initiated in the local lymph nodes (26,27). Recent observations in the course of early-phase clinical trials tend to support the notion of a potential role of T cells in the initiation and perpetuation of RA. CTLA-4Ig, which binds to the costimulatory molecules CD80 and CD86 and thus prevents their interaction with and subsequent activation of the CD28 receptor on the T cell (28), appears to be highly effective in RA. More than 50% of RA patients treated with 10 mg/kg of CTLA-4Ig achieved an ACR20 response, compared with ⬍20% of patients receiving placebo, and there was also a dose-dependent decrease in the soluble IL-2 receptor, a well-known measure of disease activity (29,30). Similar clinical effects may be attributable to inhibition of lymphocyte function–associated antigen 3–CD2 interaction, which also interferes with T cell costimulation, has significant activities in psoriatic arthritis (31), and is likely to have similar effects in RA. Thus, not only do these data suggest that T cells are involved in the pathogenesis of RA, but also they reveal that therapies directed at interfering with T cell activities may be as effective as other DMARDs. Nevertheless, these initial data will have to be confirmed in future trials. Likewise, it has to be shown that radiographic progression is halted and functional impairment significantly improved when this approach is used. A variety of therapies targeting T cells have, however, been hitherto unsuccessful (32), and the failure of T cell– directed therapies to significantly benefit more than a subset of patients suggests that perpetuation of chronic disease is multifactorial and that T cells do not represent the only responsible mechanism, at least not all of the time in every patient. Moreover, initiation of the disease process may occur prior to the first appearance of RA THERAPY AFTER TNF AND IL-1 BLOCKADE 3311 Figure 2. Expression of matrix metalloproteinases (MMPs), but not tissue inhibitor of metalloproteinases (TIMP), at the cartilage–pannus junction area of the joint from soluble human tumor necrosis factor– transgenic mice. Deparaffined joint sections were stained with anti–MMP-3 (A), anti–MMP-9 (B), anti– MMP-13 (C), or TIMP antibody (D). MMP- or TIMP-expressing cells are indicated by brown staining. JS ⫽ joint space; AC ⫽ articular cartilage; SB ⫽ subchondral bone. (Original magnification ⫻ 200.) clinical symptoms, and pathways involving the innate immune system, including dendritic cells (DCs) and macrophages, and Toll-like receptors may be operative therein (33,34). Regulatory T cells. On a more speculative basis, it is conceivable that activation of regulatory T cells, shown to be deficient in nonarthritic experimental models of autoimmune diseases (35), may also exert beneficial effects in RA. These Treg cells have been only recently described, and thus the importance of suppressor cells has been rediscovered (36). They appear to exert their down-regulating effects not via CTLA-4, but via cell–cell contact and transforming growth factor ␤ production. In experimental autoimmune models, activation of Treg cells leads to prevention and cure of disease (37). This concept has also been taken forward into human autoimmunity (38). Moreover, certain DC subsets are involved in the induction of Treg cells (33). Targets downstream of TNF and IL-1 Signal transduction cascades and the role of matrix metalloproteinases (MMPs). Once TNF or IL-1 engages its cell surface receptor, a variety of signaling pathways are triggered. Two of the most important are the NF-B and the MAPK/SAPK pathways. Both NF-B and MAPK have been shown to be activated in the RA synovial membrane (39,40). Previous observations have already revealed that stress pathways are activated in the synovium, since heat-shock protein 70 and heat-shock factor 1 are overexpressed in the lining layer and other areas (39,41). Interestingly, although TNF and IL-1 are overexpressed in the RA synovial membrane (23,41–43), and although both cytokines appear to activate the 3 MAPKs ERK, JNK, and p38 to a similar degree in vitro (40), they are differentially expressed in the synovial membrane (Figure 1). The p38 3312 REDLICH ET AL Clearly, MAPK and their downstream transcription factor, activator protein 1, are involved in the activation of different cytokine and MMP genes (44–46), although other transcription factors such as NF-B and Ets-1, involved in cytokine and MMP induction, are evidently overexpressed in the RA, but not in the osteoarthritic, synovium (39,47). In fact, an imbalance of MMPs and tissue inhibitors of metalloproteinases (TIMPs) has been Figure 3. Marked reduction of inflammation (A) and erosion (B) in the proximal interphalangeal (PIP) and metatarsophalangeal (MTP) joints of soluble human tumor necrosis factor–transgenic (hTNFtg) mice treated with an adenoviral vector for tissue inhibitor of metalloproteinases (TIMP), compared with untreated hTNFtg control (Co) mice. Bars show the mean and SEM (n ⫽ 7 in each group). ⴱ ⫽ P ⬍ 0.01 versus controls. kinase is mainly expressed in the lining layer and in endothelial cells, ERK is mainly expressed in perivascular mononuclear cells, and JNK is mainly seen in the sublining area. Since ingression of inflammatory cells into the synovial membrane via the endothelium and attachment of lining layer cells to cartilage and the cartilage–bone junction appear to be important events in the pathogenesis of RA, p38 MAPK may play a particularly important role. However, it is not clear at present which subsequent events are differentially driven by the 3 MAPKs, nor do we know why they are distinctly activated in specific regions of the synovial membrane. Figure 4. Osteoclasts are found in inflamed joints at the site of erosions in a soluble human tumor necrosis factor–transgenic (hTNFtg) mouse (A) and in a metatarsal joint of a patient with rheumatoid arthritis (RA) (B). Osteoclasts are found in the tartrate-resistant acid phosphatase (TRAP)–stained sections at the front of erosions (TRAPpositive multinucleated cells, indicated by arrows). SB ⫽ subchondral bone. (Original magnification ⫻ 200 in A; ⫻ 100 in B.) RA THERAPY AFTER TNF AND IL-1 BLOCKADE 3313 Figure 5. Evidence of severe joint inflammation in soluble human tumor necrosis factor–transgenic (hTNFtg) and c-fos⫺/⫺ hTNFtg mice, but no signs of bone erosions in the c-fos⫺/⫺ hTNFtg mouse, as demonstrated by hematoxylin and eosin–stained sections of tarsal areas of the hind paws of an hTNFtg (A), c-fos⫺/⫺ hTNFtg (B), wild-type (WT) (C), and c-fos⫺/⫺ (D) mouse. Inflammatory tissue (indicated by white arrows) and erosions (indicated by black arrows) are present in the hTNFtg mouse (A), whereas the c-fos⫺/⫺ hTNFtg mouse only shows inflammatory tissue (white arrows) but no erosions (B). The WT (C) and c-fos⫺/⫺ (D) mice lack any signs of inflammation, and therefore serve as controls. (Original magnification ⫻ 50.) observed in RA (48), and MMPs have been suggested to play a pivotal role in the cartilage and bone destruction of RA (32,49). On the basis of these findings, molecules involved in the transduction of cytokine-induced signals, such as the NF-B or p38 and JNK pathways, currently constitute important targets of therapeutic efforts, both experimentally and in early clinical trials (50,51). It will be important to reveal their degree of efficacy as well as their toxicity. In an animal model in which overexpression of TNF leads to severe destructive arthritis independent of T and B cells (52), MMPs are also highly activated, whereas TIMP is hardly found (Figure 2). Using a gene therapeutic approach, TIMP was overexpressed, and both inflammation and destruction were highly retarded in these TNF-transgenic animals (Figure 3) (48). Thus, targeting MMPs may also be of benefit in human RA. Although several xenobiotic MMP inhibitors have been tested with little success or with the occurrence of significant adverse events (32), TIMP has not yet been applied and could constitute a powerful therapy. Interestingly, the findings of the experiment demonstrated in Figure 3 suggest that in addition to gross tissue destruction, MMPs are involved in the inflammatory process via infiltration of inflammatory cells. Since, in RA, destruction correlates with the degree of joint inflammation, the above data are compatible with such notions of some linkage between inflammation and destruction in RA. Mechanisms of joint destruction. In order to pursue the question of whether or not inflammation and 3314 destruction are coupled, it is important to analyze which cells and molecules are pivotally involved in the destructive process. In recent years, it had been assumed that pannus invades cartilage and bone and that MMPs are prominently involved in the destruction of the joint (49). The synoviocytes or pannocytes, i.e., mainly the fibroblast-like synovial lining cells, have been demonstrated to be of a particularly aggressive nature (49,53– 55). On the other hand, osteoclasts have been shown to be present in RA synovial membrane near sites of erosions (56), and the factors essential for osteoclast differentiation and activation, namely receptor activator of NF-B (RANK), RANK ligand (RANKL), and macrophage colony-stimulating factor (57), are all present in the synovial membrane. Moreover, interference with the RANK/RANKL system by osteoprotegerin (OPG) leads to inhibition of destruction in adjuvant arthritis, a T cell–mediated experimental arthritis (58), and because RANK/RANKL are also important for T cell–DC interactions, it was suggested that these findings were due to blockade of T cell activities in the context of osteoclast maturation. In order to pursue this issue further, we carried out experiments utilizing the T cell– and B cell– independent TNF-transgenic animal model (59). Indeed, in this model, as in RA, osteoclasts were found at the pannus–bone junction (Figure 4). Administration of OPG to these animals significantly reduced bone erosions but did not influence inflammation (59). However, since this still did not rule out the possibility that cell–cell interactions other than the differentiation of osteoclasts were inhibited by OPG, other investigations were necessary. This finding was further expanded in a model utilizing c-Fos–deficient animals; these mice are osteopetrotic, have no overt immunologic defects, and lack osteoclasts, since c-Fos appears to be stringently required for the transduction of RANK-mediated signals (60). Crossing c-Fos–deficient mice with TNFtransgenic mice led to a form of experimental arthritis in which a high degree of inflammation and functional impairment was seen, but bone destruction was prevented (61) (Figure 5). When serum from K/BxN mice (a mouse strain with a transgenic T cell receptor that develops severe, erosive arthritis due to antibodies against glucose-6-phosphate isomerase ) was injected into osteoclast-deficient RANKL⫺/⫺ mice, similar observations were reported (63). However, in both of these animal models, cartilage matrix was degraded, as evidenced by proteoglycan loss (Figure 6). Thus, these data indicate that bone destruction in arthritis is 1) dependent on the activation of osteoclasts, REDLICH ET AL and 2) an event which can be separated from inflammation, since mechanisms that lead to the differentiation and/or activation of osteoclasts can be specifically targeted and blocked. Cartilage degradation is likely to occur because of the production of proteinases by cells within the pannus, synovial tissue, and chondrocytes, in response to several stimuli resulting from the inflammatory process. These observations on the distinctions between inflammation, cartilage degradation, and bone destruction are also indirectly supported by the finding that the highest degree of cartilage matrix degradation occurs in reactive arthritis, a disease rarely accompanied by bone erosions (64,65). However, since the degree of radiographic joint destruction correlates with bad functional outcome in RA in the long term (66), interference with the destructive process in combination with specific or nonspecific antiinflammatory measures might be an interesting and necessary therapeutic avenue. Such interference could be brought about by OPG or antibodies to RANKL (67). Targets acting in parallel or some cross-dependence with TNF and IL-1 Many cytokines other than TNF and IL-1 appear to be involved in the pathogenesis of RA. They have either an antigen-presenting cell, macrophage, or T cell origin. For example, IL-6, IL-12, IL-15, IL-17, and IL-18 all appear to be involved in RA. In some instances, these proinflammatory cytokines may be induced by TNF or IL-1 (48,68–75). Although the hierarchy of the cellular and cytokine pathways in the pathogenesis of RA is still elusive, inhibition of each of these cytokines appears to ameliorate experimental arthritis. In addition to IL-6, for which data from phase II trials are available (76) and phase III trials are ongoing, some targets, such as IL-15 and IL-18, have already been utilized in clinical trials (77). Preliminary available information suggests the efficacy of IL-15 and IL-18 blockade, but again, this effect does not appear to supercede that known for other types of interventions, further supporting the view of a redundancy and multiplicity of cells and molecules involved in the pathogenesis of RA. Another area highly neglected in recent years is autoimmunity (78). In many centers, research in RA has drifted away from its previous major focus on autoimmunity (78–80). New autoantigens have been described in recent years, and most of them have been characterized, among them RA33, Sa, BiP, and citrullinated proteins (78,81). T cell reactivity to some of these autoantigens has RA THERAPY AFTER TNF AND IL-1 BLOCKADE 3315 Figure 6. Reduction in proteoglycan (PG) content in cartilage of c-fos⫺/⫺ soluble human tumor necrosis factor–transgenic (hTNFtg) mice and receptor activator of NF-B ligand (RANKL)⫺/⫺ mice with serum transfer arthritis. Data from 2 different sets of experiments are shown. In one set (A and B), c-fos⫺/⫺ mice, which lack osteoclasts, were crossed with hTNFtg mice, which develop severe erosive arthritis, generating animals with severe inflammation but lack of osteoclasts (c-fos⫺/⫺ hTNFtg mice) (61). In another set of experiments (C–F), RANKL⫺/⫺ mice, which also lack osteoclasts, received serum from K/BxN mice, which spontaneously develop severe, autoantibody-mediated erosive arthritis, generating an immune complex–mediated arthritis model in the absence of osteoclasts (63). Toluidine blue staining of articular cartilage is shown for c-fos⫺/⫺ mice (A), c-fos⫺/⫺ hTNFtg mice (B), littermate control mice with serum transfer arthritis (C and D), and RANKL⫺/⫺ mice with serum transfer arthritis (E and F). A marked reduction in PG content, indicated by decreased staining intensity (black arrows), was found in c-fos⫺/⫺ hTNFtg mice as well as in arthritic RANKL⫺/⫺ mice and arthritic control mice. Pannus formation with associated cartilage matrix destruction and PG loss was observed in the serum transfer arthritis model (C and E). Red asterisks indicate full-thickness subchondral bone erosions with associated full-thickness cartilage destruction in control mice with serum transfer arthritis (D), which were absent in arthritic RANKL⫺/⫺ mice. (Original magnification ⫻ 200 in A and B; ⫻ 50 in C–F.) Figures 6C–F were kindly provided by Drs. Allison R. Pettit and Ellen M. Gravallese. been demonstrated in RA patients (82,83). This T cell reactivity, as well as observations on the IgG nature and the somatic mutation of such autoantibodies, clearly suggest the importance of T cell–B cell interactions in the course of the autoimmune response. Furthermore, some indications for a potential pathogenic role of autoantibodies in RA come from clinical studies in which selective blocking of B cells with anti-CD20 (rituximab) has led to a major improvement in disease activity (84,85). However, even though rheumatoid factor (RF) levels decreased or even 3316 returned to normal levels in some of these patients, the detailed mechanisms whereby anti-CD20 improves RA are currently unknown and may involve pathways other than interference with autoantibody production. Nevertheless, when trying to differentiate the characteristics of highly erosive RA from those of less erosive diseases, such as psoriatic arthritis and postinfectious arthritis, one does not find major differences in either the synovial histology or qualitative cytokine profile (86,87), but rather, a disparity is evident in the MHC associations (HLA–DR4 in RA versus class I antigens in non-RA diseases) and in the autoantibody response (RF and a variety of other autoantibodies in RA versus none in the other disorders). In fact, these other arthritides are subsumed as “seronegative” arthritides. Moreover, RF has been shown to be associated with more aggressive RA (88). Although this has also been suggested for the shared epitope of the HLA–DR cluster, that association is possibly due to an association of HLA–DR4 with RF (89). Thus, the presence of autoantibodies appears to sustain arthritic destruction. This has also been seen in experimental models of arthritis in which autoantibodies were found when destruction occurred, including the model of TNF-transgenic mice bearing a primary non– immune-mediated arthritic disease (48). On the other hand, autoantibodies could be a consequence of inflammation and destruction, and this is suggested by the occurrence of RF in the course of infections and chronic diseases as well as by observations in TNF-transgenic mice in which inhibition of inflammation and/or destruction prevented the generation of autoantibodies (48). This is further supported by clinical studies in which, following successful therapy, RF titers decreased (6,90). How could autoantibodies be related to joint destruction? It is conceivable that Fc receptor and/or complement or complement receptor–mediated events play an important role in this respect. In fact, it has long been known that complement is activated in RA synovial fluid, but not in most other diseases (91). Moreover, mice deficient in either Fc␥ receptors or complement receptors are protected from at least some forms of experimental arthritis (92–94). Taken together, these data suggest that immune complexes, presumably consisting of autoantibodies and autoantigens, may promote inflammation and tissue destruction in RA. The ability of immune complexes to induce TNF production by macrophages has previously been determined (95). Thus, since autoimmune phenomena are clearly associated with the destructiveness of arthritis, one could REDLICH ET AL hypothesize that Fc-receptor– and/or complementreceptor–mediated signaling could be operative in the differentiation and activation of osteoclasts in conjunction with or via enhanced induction of TNF and/or RANKL or RANK. All of these observations suggest that targeting Fc-receptor function and complement components may constitute interesting therapeutic avenues for RA. Conclusion Current therapies for RA are highly efficacious but fail to induce remission in most patients (2–6,11,13). On the basis of the known linkage of multiple pathogenetic pathways to arthritis, it must be assumed that targeting of individual cells or molecules is insufficient. This is suggested in even apparently straightforward mouse models, as in the TNF-transgenic mouse model in which TNF blockade is insufficient to abrogate inflammation or destruction and combination therapy is much more efficacious (96). Obviously, this notion should not impede our search for individual novel targets and individual novel agents (97), which are still much needed for the treatment of RA. However, drawing conclusions from the above observations is necessary for the design of new combination strategies, such as inhibition of TNF plus at least 1 additional target. In this context, it must be our aim to interfere with both the pathways that lead to inflammation and those that ultimately activate the destruction cascade, although it is possible that complete inhibition of the inflammatory response will also prevent bone and cartilage destruction. It is this aim that we must keep in mind when designing new strategies and new clinical trials in the era after TNF and IL-1 blockade. REFERENCES 1. Elliott MJ, Maini RN, Feldmann M, Kalden JR, Antoni C, Smolen JS, et al. Randomised double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha (cA2) versus placebo in rheumatoid arthritis. Lancet 1994;344:1105–10. 2. Bathon JM, Martin RW, Fleischmann RM, Tesser JR, Schiff MH, Keystone EC, et al. A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis. N Engl J Med 2000; 343:1586–93. 3. Bresnihan B, Alvaro-Gracia JM, Cobby M, Doherty M, Domljan Z, Emery P, et al. Treatment of rheumatoid arthritis with recombinant human interleukin-1 receptor antagonist. Arthritis Rheum 1998;41:2196–204. 4. Lipsky PE, van der Heijde DM, St Clair EW, Furst DE, Breedveld FC, Kalden JR, et al, the Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. Infliximab and methotrexate in the treatment of rheumatoid arthritis. N Engl J Med 2000;343:1594–602. RA THERAPY AFTER TNF AND IL-1 BLOCKADE 5. Weinblatt ME, Keystone EC, Furst DE, Moreland LW, Weisman MH, Birbara CA, et al. Adalimumab, a fully human anti–tumor necrosis factor ␣ monoclonal antibody, for the treatment of rheumatoid arthritis in patients taking concomitant methotrexate: The ARMADA trial. Arthritis Rheum 2003;48:35–45. 6. Smolen JS, Kalden JR, Scott DL, Rozman B, Kvien TK, Larsen A, et al. 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. 7. Arend WP, Dayer JM. Cytokines and cytokine inhibitors or antagonists in rheumatoid arthritis. Arthritis Rheum 1990;33: 305–15. 8. Feldmann M, Maini RN. Anti-TNF alpha therapy of rheumatoid arthritis: what have we learned? Annu Rev Immunol 2001;19: 163–96. 9. Genovese MC, Bathon JM, Martin RW, Fleischmann RM, Tesser JR, Schiff MH, et al. Etanercept versus methotrexate in patients with early rheumatoid arthritis: two-year radiographic and clinical outcomes. Arthritis Rheum 2002;46:1443–50. 10. Furst DE, Breedveld FC, Kalden JR, Smolen JS, Antoni CE, Bijlsma JW, et al. Updated consensus statement on biological agents for the treatment of rheumatoid arthritis and other rheumatic diseases (May 2002). Ann Rheum Dis 2002;61 Suppl 2:ii2–7. 11. 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. 12. Schiff M, Furst D, Cohen M, Fleischmann R, Macri M, Joh TS, et al. The long-term safety of pegylated recombinant methionyl human soluble tumor necrosis factor type I (PEG STNF-RI): an extension study for rheumatoid arthritis patients completing previous PEG STNF-RI clinical studies. Arthritis Rheum 2001;44 Suppl 9:S158. 13. Cohen S, Hurd E, Cush J, Schiff M, Weinblatt ME, Moreland LW, 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. 14. Felson DT, Anderson JJ, Boers M, Bombardier C, Chernoff M, Fried B, et al. The American College of Rheumatology preliminary core set of disease activity measures for rheumatoid arthritis clinical trials. Arthritis Rheum 1993;36:729–40. 15. Van Gestel AM, Prevoo MLL, Van ’t Hof MA, van Rijswijk MH, van de Putte LBA, van Riel PLCM. Development and validation of the European League Against Rheumatism response criteria for rheumatoid arthritis: comparison with the preliminary American College of Rheumatology and the World Health Organization/International League Against Rheumatism Criteria. Arthritis Rheum 1996;39:34–40. 16. Van Gestel AM, van Riel PLCM. Validation of rheumatoid arthritis improvement criteria that include simplified joint counts. Arthritis Rheum 1998;41:1845–50. 17. Antoni C, Kavanaugh A, Manger B, Kalden J, Keenan G, Schaible T, et al. Responses to infliximab therapy in the ATTRACT trial assessed with the Disease Activity Score (DAS): clinical response measured by DAS correlates with arrest of radiologic progression and shows higher response rates than ACR20 criteria. Arthritis Rheum 2000;43 Suppl 9:S961. 18. Aletaha D, Smolen JS. The rheumatoid arthritis patient in the clinic: comparing more than 1300 consecutive DMARD courses. Rheumatology (Oxford) 2002;41:1367–74. 19. Van Den Berg WB. Uncoupling of inflammatory and destructive mechanisms in arthritis. Semin Arthritis Rheum 2001;30 Suppl 2:7–16. 20. Ishihara K, Hirano T. IL-6 in autoimmune disease and chronic 3317 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. inflammatory proliferative disease. Cytokine Growth Factor Rev 2002;13:357–68. Firestein GS, Zvaifler NJ. How important are T cells in chronic rheumatoid synovitis? II. T cell–independent mechanisms from beginning to end. Arthritis Rheum 2002;46:298–308. Gregersen PK, Silver J, Winchester RJ. The shared epitope hypothesis: an approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis. Arthritis Rheum 1987;30: 1205–13. Steiner G, Tohidast AM, Witzmann G, Vesely M, Studnicka BA, Gal A, et al. Cytokine production by synovial T cells in rheumatoid arthritis. Rheumatology (Oxford) 1999;38:202–13. Tugwell P, Pincus T, Yocum D, Stein M, Gluck O, Kraag G, et al, The Methotrexate-Cyclosporine Combination Study Group. Combination therapy with cyclosporine and methotrexate in severe rheumatoid arthritis. N Engl J Med 1995;333:137–41. Burger D, Dayer JM. Cytokines, acute-phase proteins, and hormones: IL-1 and TNF-alpha production in contact-mediated activation of monocytes by T lymphocytes. Ann N Y Acad Sci 2002;966:464–73. Scheinecker C, McHugh R, Shevach EM, Germain RN. Constitutive presentation of a natural tissue autoantigen exclusively by dendritic cells in the draining lymph node. J Exp Med 2002;196: 1079–90. Mandik-Nayak L, Wipke BT, Shih FF, Unanue ER, Allen PM. Despite ubiquitous autoantigen expression, arthritogenic autoantibody response initiates in the local lymph node. Proc Natl Acad Sci U S A 2002;99:14368–73. Slavik JM, Hutchcroft JE, Bierer BE. CD28/CTLA-4 and CD80/ CD86 families: signaling and function. Immunol Res 1999;19:1–24. Moreland LW, Alten R, Van den Bosch F, Appelboom T, Leon M, Emery P, et al. Costimulatory blockade in patients with rheumatoid arthritis: a pilot, dose-finding, double-blind, placebo-controlled clinical trial evaluating CTLA-Ig and LE429Y eighty-five days after the first infusion. Arthritis Rheum 2002;46:1470–9. Kremer J, Westhovens R, Leon M, Di Georgio E, Alten R, Steinfeld S, et al. A phase 2B multi-center, randomized, doubleblind, placebo controlled study to evaluate the safety and efficacy of two different doses of CTLA4Ig administered intravenously to subjects with active rheumatoid arthritis while receiving methotrexate. Arthritis Rheum 2002;46 Suppl 9:S203. Kraan MC, van Kuijk AW, Dinant HJ, Goedkoop AY, Smeets TJ, de Rie MA, et al. Alefacept treatment in psoriatic arthritis: reduction of the effector T cell population in peripheral blood and synovial tissue is associated with improvement of clinical signs of arthritis. Arthritis Rheum 2002;46:2776–84. Keystone E. Treatments no longer in development for rheumatoid arthritis. Ann Rheum Dis 2002;61 Suppl 2:ii43–5. Thomas R, Thompson A, Martin E, O’Sullivan B. Dendritic cells. In: Smolen JS, Lipsky PE, editors. Targeted therapies in rheumatology. London-New York: Martin Dunitz; 2003. p. 61–81. Krieg AM. A role for Toll in autoimmunity. Nat Immunol 2002;3:423–4. Shevach EM. CD4⫹ CD25⫹ suppressor T cells: more questions than answers. Nat Rev Immunol 2002;2:389–400. Chatenoud L, Salomon B, Bluestone JA. Suppressor T cells: they’re back and critical for regulation of autoimmunity! Immunol Rev 2001;182:149–63. Suri-Payer E, Amar AZ, Thornton AM, Shevach EM. CD4⫹CD25⫹ T cells inhibit both the induction and effector function of autoreactive T cells and represent a unique lineage of immunoregulatory cells. J Immunol 1998;160:1212–8. Horwitz DA, Gray JD, Zheng SG. The potential of human regulatory T cells generated ex vivo as a treatment for lupus and other chronic inflammatory diseases. Arthritis Res 2002;4:241–6. Han Z, Boyle DL, Manning AM, Firestein GS. AP-1 and NF- 3318 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. kappaB regulation in rheumatoid arthritis and murine collageninduced arthritis. Autoimmunity 1998;28:197–208. Schett G, Tohidast AM, Smolen JS, Schmid BJ, Steiner CW, Bitzan P, et al. Activation, differential localization, and regulation of the stress-activated protein kinases, extracellular signal–regulated kinase, c-Jun N-terminal kinase, and p38 mitogen-activated protein kinase, in synovial tissue and cells in rheumatoid arthritis. Arthritis Rheum 2000;43:2501–12. Schett G, Redlich K, Xu Q, Bizan P, Groger M, Tohidast-Akrad M, et al. Enhanced expression of heat shock protein 70 (hsp70) and heat shock factor 1 (HSF1) activation in rheumatoid arthritis synovial tissue: differential regulation of hsp70 expression and hsf1 activation in synovial fibroblasts by proinflammatory cytokines, shear stress, and antiinflammatory drugs. J Clin Invest 1998;102: 302–11. Deleuran BW, Chu CQ, Field M, Brennan FM, Katsikis P, Feldmann M, et al. Localization of interleukin-1 alpha, type 1 interleukin-1 receptor and interleukin-1 receptor antagonist in the synovial membrane and cartilage/pannus junction in rheumatoid arthritis. Br J Rheumatol 1992;31:801–9. Koch AE, Kunkel SL, Chensue SW, Haines GK, Strieter RM. Expression of interleukin-1 and interleukin-1 receptor antagonist by human rheumatoid synovial tissue macrophages. Clin Immunol Immunopathol 1992;65:23–9. Saklatvala J, Dean J, Finch A. Protein kinase cascades in intracellular signalling by interleukin-I and tumour necrosis factor. Biochem Soc Symp 1999;64:63–77. Van den Berghe W, Vermeulen L, De Wilde G, De Bosscher K, Boone E, Haegeman G. Signal transduction by tumor necrosis factor and gene regulation of the inflammatory cytokine interleukin-6. Biochem Pharmacol 2000;60:1185–95. Karin M, Lin A. NF-kappaB at the crossroads of life and death. Nat Immunol 2002;3:221–7. Redlich K, Kiener HP, Schett G, Tohidast-Akrad M, Selzer E, Radda I, et al. Overexpression of transcription factor Ets-1 in rheumatoid arthritis synovial membrane: regulation of expression and activation by interleukin-1 and tumor necrosis factor ␣. Arthritis Rheum 2001;44:266–74. Schett G, Hayer S, Tohidast-Akrad M, Schmid BJ, Lang S, Turk B, et al. Adenovirus-based overexpression of tissue inhibitor of metalloproteinases 1 reduces tissue damage in the joints of tumor necrosis factor ␣ transgenic mice. Arthritis Rheum 2001;44: 2888–98. Harris ED Jr. Rheumatoid arthritis: pathophysiology and implications for therapy. N Engl J Med 1990;322:1277–89. Firestein GS, Manning AM. Signal transduction and transcription factors in rheumatic disease. Arthritis Rheum 1999;42:609–21. Orchard S. Kinases as targets: prospects for chronic therapy. Curr Opin Drug Discov Devel 2002;5:713–7. Keffer J, Probert L, Cazlaris H, Georgopoulos S, Kaslaris E, Kioussis D, et al. Transgenic mice expressing human tumour necrosis factor: a predictive genetic model of arthritis. EMBO J 1991;10:4025–31. Pap T, Claus A, Ohtsu S, Hummel KM, Schwartz P, Drynda S, et al. Osteoclast-independent bone resorption by fibroblast-like cells. Arthritis Res Ther 2003;5:R163–R173. Pap T, van der Laan WH, Aupperle KR, Gay RE, Verheijen JH, Firestein GS, et al. Modulation of fibroblast-mediated cartilage degradation by articular chondrocytes in rheumatoid arthritis. Arthritis Rheum 2000;43:2531–36. Sack U, Kuhn H, Ermann J, Kinne RW, Vogt S, Jungmichel D, et al. Synovial tissue implants from patients with rheumatoid arthritis cause cartilage destruction in knee joints of SCID.bg mice. J Rheumatol 1994;21:10–6. Gravallese EM, Harada Y, Wang JT, Gorn AH, Thornhill TS, Goldring SR. Identification of cell types responsible for bone REDLICH ET AL 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. resorption in rheumatoid arthritis and juvenile rheumatoid arthritis. Am J Pathol 1998;152:943–51. Teitelbaum SL. Bone resorption by osteoclasts. Science 2000;289: 1504–8. Kong YY, Feige U, Sarosi I, Bolon B, Tafuri A, Morony S, et al. Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 1999; 402:304–9. Redlich K, Hayer S, Maier A, Dunstan CR, Tohidast-Akrad M, Lang S, et al. Tumor necrosis factor ␣–mediated joint destruction is inhibited by targeting osteoclasts with osteoprotegerin. Arthritis Rheum 2002;46:785–92. Grigoriadis AE, Wang ZQ, Cecchini MG, Hofstetter W, Felix R, Fleisch HA, et al. c-Fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science 1994;266: 443–8. Redlich K, Hayer S, Ricci R, David JP, Tohidast-Akrad M, Kollias G, et al. Osteoclasts are essential for TNF-alpha-mediated joint destruction. J Clin Invest 2002;110:1419–27. Matsumoto I, Maccioni M, Lee DM, Maurice M, Simmons B, Brenner M, et al. How antibodies to a ubiquitous cytoplasmic enzyme may provoke joint-specific autoimmune disease. Nat Immunol 2002;3:360–5. Pettit AR, Ji H, von Stechow D, Muller R, Goldring SR, Choi Y, et al. TRANCE/RANKL knockout mice are protected from bone erosion in a serum transfer model of arthritis. Am J Pathol 2001;159:1689–99. Saxne T, Heinegard D. Cartilage oligomeric matrix protein: a novel marker of cartilage turnover detectable in synovial fluid and blood. Br J Rheumatol 1992;31:583–91. Saxne T, Glennas A, Kvien TK, Melby K, Heinegard D. Release of cartilage macromolecules into the synovial fluid in patients with acute and prolonged phases of reactive arthritis. Arthritis Rheum 1993;36:20–5. Welsing PM, van Gestel AM, Swinkels HL, Kiemeney LA, van Riel PL. The relationship between disease activity, joint destruction, and functional capacity over the course of rheumatoid arthritis. Arthritis Rheum 2001;44:2009–17. Bolon B, Shalhoub V, Kostenuik PJ, Campagnuolo G, Morony S, Boyle WJ, et al. Osteoprotegerin, an endogenous antiosteoclast factor for protecting bone in rheumatoid arthritis. Arthritis Rheum 2002;46:3121–35. Nishimoto N, Yoshizaki K, Kishimoto T. Interleukin-6. In: Smolen JS, Lipsky PE, editors. Targeted therapies in rheumatology. London-New York: Martin Dunitz; 2003. p. 231–41. Gracie JA, Forsey RJ, Chan WL, Gilmour A, Leung BP, Greer MR, et al. A proinflammatory role for IL-18 in rheumatoid arthritis. J Clin Invest 1999;104:1393–401. Chabaud M, Durand JM, Buchs N, Fossiez F, Page G, Frappart L, et al. Human interleukin-17: a T cell–derived proinflammatory cytokine produced by the rheumatoid synovium. Arthritis Rheum 1999;42:963–70. McInnes IB, Leung BP, Sturrock RD, Field M, Liew FY. Interleukin-15 mediates T cell-dependent regulation of tumor necrosis factor-alpha production in rheumatoid arthritis. Nat Med 1997;3: 189–95. Joosten LA, Heuvelmans-Jacobs M, Lubberts E, van De Loo FA, Bakker AC, Helsen MM, et al. Local interleukin-12 gene transfer promotes conversion of an acute arthritis to a chronic destructive arthritis. Arthritis Rheum 2002;46:1379–89. Dinarello CA. Interleukin-18. In: Smolen JS, Lipsky PE, editors. Targeted therapies in rheumatology. London-New York: Martin Dunitz; 2003. p. 277–96. McInnes IB. Interleukin-15. In: Smolen JS, Lipsky PE, editors. Targeted therapies in rheumatology. London-New York: Martin Dunitz; 2003. p. 259–75. Moriguchi M. Interleukin-12, interleukin-23 and interleukin 27. RA THERAPY AFTER TNF AND IL-1 BLOCKADE 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. In: Smolen JS, Lipsky PE, editors. Targeted therapies in rheumatology. London-New York: Martin Dunitz; 2003. p. 243–57. Choy EH, Isenberg DA, Garrood T, Farrow S, Ioannou Y, Bird H, et al. Therapeutic benefit of blocking interleukin-6 activity with an anti–interleukin-6 receptor monoclonal antibody in rheumatoid arthritis: a randomized, double-blind, placebo-controlled, doseescalation trial. Arthritis Rheum 2002;46:3143–50. Liew FY, McInnes IB. Role of interleukin 15 and interleukin 18 in inflammatory response. Ann Rheum Dis 2002;61 Suppl 2:ii100–2. Smolen JS, Steiner G. Rheumatoid arthritis is more than cytokines: autoimmunity and rheumatoid arthritis. Arthritis Rheum 2001;44:2218–20. Pope RM, Talal N. Autoimmunity in rheumatoid arthritis. Concepts Immunopathol 1985;1:219–50. Williams R-CJ. Autoimmune mechanisms involved in the pathogenesis of rheumatoid arthritis. Adv Dent Res 1996;10:47–51. Schellekens GA, Visser H, de Jong BA, van den Hoogen FH, Hazes JM, Breedveld FC, et al. The diagnostic properties of rheumatoid arthritis antibodies recognizing a cyclic citrullinated peptide. Arthritis Rheum 2000;43:155–63. Fritsch R, Eselbock D, Skriner K, Jahn-Schmid B, Scheinecker C, Bohle B, et al. Characterization of autoreactive T cells to the autoantigens heterogeneous nuclear ribonucleoprotein A2 (RA33) and filaggrin in patients with rheumatoid arthritis. J Immunol 2002;169:1068–76. Blass S, Union A, Raymackers J, Schumann F, Ungethum U, Muller-Steinbach S, et al. The stress protein BiP is overexpressed and is a major B and T cell target in rheumatoid arthritis. Arthritis Rheum 2001;44:761–71. Leandro MJ, Edwards JC, Cambridge G. Clinical outcome in 22 patients with rheumatoid arthritis treated with B lymphocyte depletion. Ann Rheum Dis 2002;61:883–8. De Vita S, Zaja F, Sacco S, De Candia A, Fanin R, Ferraccioli G. Efficacy of selective B cell blockade in the treatment of rheumatoid arthritis: evidence for a pathogenetic role of B cells. Arthritis Rheum 2002;46:2029–33. 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. Smith MD, O’Donnell J, Highton J, Palmer DG, Rozenbilds M, Roberts-Thomson PJ. Immunohistochemical analysis of synovial membranes from inflammatory and non-inflammatory arthritides: 3319 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. scarcity of CD5 positive B cells and IL2 receptor bearing T cells. Pathology 1992;24:19–26. Scott DL, Symmons DP, Coulton BL, Popert AJ. Long-term outcome of treating rheumatoid arthritis: results after 20 years. Lancet 1987;1:1108–11. Harrison B, Thomson W, Symmons D, Ollier B, Wiles N, Payton T, et al. The influence of HLA–DRB1 alleles and rheumatoid factor on disease outcome in an inception cohort of patients with early inflammatory arthritis. Arthritis Rheum 1999;42:2174–83. Maini R, St Clair EW, Breedveld F, Furst D, Kalden J, Weisman M, et al, 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. Ruddy S, Austen KF. Activation of the complement and properdin systems in rheumatoid arthritis. Ann NY Acad Sci 1975;256: 96–104. Grant EP, Picarella D, Burwell T, Delaney T, Croci A, Avitahl N, et al. Essential role for the C5a receptor in regulating the effector phase of synovial infiltration and joint destruction in experimental arthritis. J Exp Med 2002;196:1461–71. Kleinau S, Martinsson P, Heyman B. Induction and suppression of collagen-induced arthritis is dependent on distinct Fcgamma receptors. J Exp Med 2000;191:1611–6. Van Lent PL, Nabbe K, Blom AB, Holthuysen AE, Sloetjes A, van de Putte LB, et al. Role of activatory Fc gamma RI and Fc gamma RIII and inhibitory Fc gamma RII in inflammation and cartilage destruction during experimental antigen-induced arthritis. Am J Pathol 2001;159:2309–20. Debets JM, Van de Winkel JG, Ceuppens JL, Dieteren IE, Buurman WA. Cross-linking of both Fc gamma RI and Fc gamma RII induces secretion of tumor necrosis factor by human monocytes, requiring high affinity Fc–Fc gamma R interactions: functional activation of Fc gamma RII by treatment with proteases or neuraminidase. J Immunol 1990;144:1304–10. Zwerina J, Hayer S, Tohidast-Akrad M, Bergmeister H, Redlich K, Feige U, et al. Single and combined inhibition of tumor necrosis factor, interleukin-1, and RANKL pathways in tumor necrosis factor–induced arthritis: effects on synovial inflammation, bone erosion, and cartilage destruction. Arthritis Rheum. In press. Smolen JS, Steiner G. Therapeutic strategies for rheumatoid arthritis. Nat Rev Drug Discov 2003;2:473–88.