Anti-TNF Therapy

153

REVIEW

Anti-TNF Therapy for Rheumatoid Arthritis and Other Inflammatory Diseases Peter C. Taylor

Abstract The availability of agents that block the biological activity of tumor necrosis factor α (TNFα) in rheumatoid arthritis (RA) has permitted studies that confirm the key role of this cytokine in the pathogenesis of this disease. To date, two anti-TNF agents, infliximab and etanercept, have been approved for use in treatment. Clinical trials of these agents demonstrate efficacy for the control of symptoms and signs and acceptable safety in patients who have failed to respond adequately to conventional therapy. Combination with methotrexate appears to be particularly effective and may provide the main initial indication for clinical application in the first instance. Repeated administration of anti-TNF therapies over a one year period results in sustained reduction in symptoms and signs of RA in the majority of patients. It has recently become apparent that anti-TNF therapy protects joints from structural damage. These findings imply that TNFα has a critical role in the bone and cartilage damage associated with RA. Evidence to date support the hypothesis that there are 2 particularly important mechanisms of action; deactivation of the proinflammatory cytokine cascade at the site of inflammation and diminished recruitment of inflammatory cells from blood to the rheumatoid joint. Index Entries: Rheumatoid arthritis; tumor necrosis factor α (TNF); anti-TNF therapy; monoclonal antibodies; fusion protein.

many cytokines, for example, interferonγ (IFN-γ), with chiefly proinflammatory activity can also in some instances have antiinflammatory properties. Similarly, cytokines with predominantly antiinflammatory activity, such as IL-10 and TGFβ, may also exhibit proinflammatory properties and therefore have pathogenic potential. Paracrine or autocrine pathways involving cytokines with either pro- or antiinflammatory activity can lead to reverberating networks determining whether chronic inflammation results.

1. Introduction Cytokines are small proteins and major mediators of local intercellular communication required for an integrated response to a variety of stimuli in immune and inflammatory responses. By binding their cognate receptors on target cells these short-lived molecules play a role in many important biological processes including cell proliferation, activation, death, and differentiation. During an inflammatory response many cytokines are synthesized by a wide range of cell types including leukocytes and fibroblasts. Some cytokines are proinflammatory, such as interleukin-1 (IL-1) and tumor necrosis factor α (TNFα); others such as interleukin-10 (IL-10) and transforming growth factor β (TGFβ) exert predominantly antiinflammatory effects. However, it is now understood that

1.1. Cytokine Antagonists as Therapeutic Agents and Probes of Pathogenesis The availability of specific high affinity antagonists to proinflammatory cytokines affords the opportunity to evaluate any potential therapeutic

*Author to whom all correspondence and reprint requests should be addressed: Peter C. Taylor MA, PhD, FRCP, Senior Lecturer and Honorary Consultant Rheumatologist, The Kennedy Institute Division, Imperial College School of Medicine, 1 Aspenlea Road, London, W6 8LH. E-mail: [email protected] Molecular Biotechnology 2001 Humana Press Inc. All rights of any nature whatsoever reserved. 1073–6085/2001/19:2/153–168/$14.00

MOLECULAR BIOTECHNOLOGY

153

Volume 19, 2001

154 benefits in a pathological setting, although also investigating the biology of a given cytokine by means of specific blockade in vivo. The first demonstration of the efficacy of specific cytokine blockade in the treatment of human chronic inflammatory disease was the use of chimeric anti-TNFα monoclonal antibodies (mAbs) to treat patients with rheumatoid arthritis (RA) (1). Other biological agents employed with a view to inhibit proinflammatory cytokine activity in RA include soluble cytokine receptors, cytokine receptor antagonists, fusion proteins combining cytokines or soluble cytokine receptors with human Fc constructs, or polyethylene glycol and counter-regulatory cytokines which oppose actions of the targets of cytokines (i.e., IL-10 or IL-4 opposing the effects of TNFα and IL-1). Means to inhibit processing or synthesis of IL-1 or TNF are also under consideration. In this article, I shall focus on the use of antibodies to cytokines in the treatment of RA, with particular emphasis on agents targeting TNFα.

2. Rheumatoid Arthritis Rheumatoid arthritis (RA) is a common human disease with a prevalence of about 1%. The clinical presentation is heterogeneous with a wide spectrum of age of onset, degree of joint involvement, and severity. At the onset of symptoms it is difficult to predict which patients will follow a more severe disease course. In past decades the treatment strategy for RA has been based on the premise that the disease prognosis is generally favorable. However, the majority of patients with a more aggressive disease evolution become clinically disabled within 20 yr and for those with severe disease or extra-articular features the mortality is equivalent to that of patients with three vessel coronary artery disease or stage IV Hodgkin’s lymphoma (2) thus, the notion that RA is a benign disease has been discredited. The traditional treatment of RA is represented by a pyramidal approach starting with nonsteroidal antiinflammatory drugs and progressing to so-called disease modifying agents such as gold, sulphasalazine, and methotrexate (MTX). However,

MOLECULAR BIOTECHNOLOGY

Taylor up to 90% of patients with aggressive synovitis have radiological evidence of bone erosion within 2 yr of diagnosis, despite treatment (3). Such observations have led to a recent trend toward much earlier use of disease modifying antirheumatic drugs in the pharmacological management of disease. Recent therapeutic advances have accompanied progress in defining the pathogenesis of RA. Although there is abundant evidence that RA is an immune-mediated disease, there are many unresolved questions, for example, is RA primarily an autoimmune disease; is the initiating agent infectious, a self antigen or both; and to what extent does the course of the disease depend on systemic or joint specific events? RA is characterized by chronic inflammation of synovial joints with synovial proliferation and infiltration by blood derived cells, in particular, memory T cells, macrophages, and plasma cells, all of which show signs of activation (4,5). The augmented cell mass is sustained by prominent development of new blood vessels. In the chronic phase of disease, capillaries and post capillary venules are particularly evident in the sublining region, histological sections sometimes demonstrating mononuclear and polymorphonuclear leukocytes in close aposition to vascular endothelium and probably in the process of margination and adhesion prior to migration into the inflamed tissue. Synovium becomes markedly hyperplastic and locally invasive at the interface of cartilage and bone with progressive destruction of these tissues in the majority of cases. This invasive tissue is referred to as “pannus,” comprising mainly lining cells with the appearance of proliferating mesenchymal cells with very little sublining lymphocytic infiltration. The accompanying destruction of bone and cartilage is thought to be principally mediated by cytokine induced degradative enzymes, especially matrix metalloproteinases. In most cases the principal manifestation of rheumatoid disease is in the synovial joint (Fig. 1), but there is also evidence of systemic involvement, for example the upregulation of acute phase proteins, and in severe cases, involvement of other organs.

Volume 19, 2001

Anti-TNF Therapy

Fig. 1. Synovial joint in health and rheumatoid arthritis indicating cellular components and sites of destruction in diseased joint.

2.1. Role of Cytokines in the Pathogenesis of RA Cytokines derived from macrophages and fibroblasts are abundant in the rheumatoid synovium. These include IL-1, TNFα, Granulocytemacrophage colony stimulating factor (GM-CSF), interleukin-6 (IL-6) and numerous chemo-attractant cytokines known as chemokines. Many of these factors are of importance in regulating inflammatory cell migration and activation. By contrast, given the extent of synovial inflammation and lymphocytic infiltration, factors produced by T cells, for example, IFNγ, interleukin-2 (IL-2), and interleukin-4 (IL-4) are surprisingly sparsely expressed (6,7). Studies from a number of laboratories have confirmed the expression of this extensive range of proinflammatory cytokines in human synovial tissue samples regardless of differences in donor MOLECULAR BIOTECHNOLOGY

155 disease duration, severity or even drug therapy (8). Such observations imply prolonged cytokine expression in rheumatoid synovium, contrasting with the transient production induced by mitogenic stimulation. This hypothesis was confirmed by the finding of proinflammatory cytokine production in dissociated RA synovial membrane cell cultures over several days in the absence of extrinsic stimulation (9). This observation suggested the presence of signals regulating prolonged cytokine synthesis within the RA synovial membrane cell cultures. A key observation, in the face of a heterogeneous population of cells producing numerous cytokine and other noncytokine messengers, was that in RA synovial cell cultures, addition of anti-TNF antibodies strikingly reduced the production of other proinflammatory cytokines, including IL-1, GM-CSF, IL-6, and IL-8 (10). Furthermore, using the same RA synovial cell culture system, blockade of IL-1 by means of the IL-1 receptor antagonist results in reduced IL-6 and IL-8 production but not that of TNFα (11). These observations support the concept that TNFα occupies the dominant position at the apex of a proinflammatory cytokine network (Fig. 2). As a pleiotropic cytokine with the ability to enhance synovial proliferation and production of prostaglandins, metalloproteinases (12), and cytokines, TNFα was seen as representing a potential therapeutic target in RA. In the chronic inflammatory situation in rheumatoid synovium there is also upregulation of multiple antiinflammatory mediators but at a level insufficient to suppress synovitis. There is abundant expression of IL-10 (13) and TGFβ both in latent and active form (14,15). Naturally occurring cytokine inhibitors, such as interleukin-1 receptor antagonists (IL-1ra) and soluble TNF receptors, the specific inhibitors of IL-1 and TNFα respectively, (16) are also upregulated in the rheumatoid joint.

3. TNFα as a Mediator of Disease As long ago as 1893 it was noted that severe infection may lead to reduction in the bulk of a malignant tumor (17), an observation now attributed to infection induced release of cytotoxic Volume 19, 2001

156

Taylor Jarisch-Herxheimer reaction (29) and subsequent to infusion of anti-CD3 monoclonal antibodies for the treatment of acute graft rejection (30). It might be predicted that TNFα blockade would be of therapeutic benefit in several of these disease states.

3.1. Strategies to Block TNFα

Fig. 2. Cytokine cascade in rheumatoid arthritis.

cytokines such as TNFα. TNFα was purified (18) and the gene encoding it cloned (19–21) in the mid 1980s. Since then many biological properties have been reported in addition of the induction of cachectia and tumor lysis that led to its discovery (22) (Fig. 3). TNFα is synthesized as a protein with a molecular weight of 26 kDa and comprising 233 amino acids. This proprotein is cleaved by TNFα converting enzyme (a specific metalloproteinase) yielding a monomeric 17 kDa molecule comprising 157 amino acids. Under physiological conditions, TNFα exists as a noncovalently bound, cone shaped homotrimer (23). The biological activity of TNFα is mediated by crosslinking of membrane bound receptors which exist in 2 isoforms; TNF receptor I (TNFRI, p55) and TNF receptor II (TNFRII, p75) (24). The extracellular domains of both receptors can be shed such that the soluble forms retain the ability to bind TNFα, thus limiting acute TNFα induced effects (25,26). It is not possible to detect TNFα in the peripheral blood of healthy humans (27) even by means of very sensitive assays (with a 200 attomolar detection limit equivalent to 12 × 105 TNF trimers in 1 mL of plasma) based on the extremely high binding specificity and affinity of p55 TNF receptor for TNFα. By contrast, TNFα can sometimes be detected in up to nanomolar concentrations in a variety of infectious diseases and inflammatory diseases including acute septic shock, bacterial meningitis, cerebral malaria, adult respiratory distress syndrome, AIDS, Crohns disease, and RA (28). High concentrations of plasma TNFα also occur in therapy associated syndromes such as MOLECULAR BIOTECHNOLOGY

The effects of released TNFα can be antagonized by anti-TNFα antibodies or soluble TNFα receptors. A number of other endogenous or synthetic anti-TNFα agents inhibit synthesis of the molecule. Endogenous agents include cytokines such as IL-4 (31), IL-10 (32), and TGFβ (33) in addition to noncytokine mediators such as corticosteroids (34), adenosine (35,36), histamine (37), and nitric oxide (38,39). Drugs may interfere with TNF production at a number of points in the synthetic pathway. Thus TNFα transcription is suppressed by pentoxifyline, a phosphodiesterase inhibitor, and TNFα translation is inhibited by dexamethasone (34). The half-life of TNFα mRNA is decreased by thalidomide (40).

4. Anti-Cytokine Antibodies as Therapeutic Agents: Clinical Research and Trials

4.1. Chimeric Monoclonal Anti-TNFα Antibody (infliximab) in the Treatment of Rheumatoid Arthritis Monoclonal antibodies that block TNFα prevent the development of symmetrical, chronic inflammatory polyarthritis, resembling RA, in transgenic mice that express large amounts of TNFα (41). Furthermore, in CIA, anti-TNFα monoclonal antibody (mAb) ameliorate established joint disease when administered in multiple doses of 10–12 mg/kg (42). The use of anti-TNFα monoclonal antibodies in human RA patients has provided the first evidence that TNFα is a potential therapeutic target in this disease. It has also provided a means for human in vivo testing of the hypothesis, formulated on the basis of in vitro studies, namely that in RA, TNFα is an important pathogenic agent and regulator of other proinflammatory mediators. The first antibody to be used in clinical trials was cA2, or infliximab, (from Centocor, Malvern, PA), a chimerized (human IgG1/mouse Fv) mAb Volume 19, 2001

Anti-TNF Therapy

157

Fig. 3. The biological activities of TNFα.

shown to bind TNFα with high affinity (1.8 × 10–9 kDa) and neutralizing various biological activities in vitro (43). The rationale for the use of chimerized antibodies was twofold; an attempt to reduce immunogenicity as compared to a murine monoclonal antibody (infliximab is 75% human immunoglobulin) and to increase the half-life of murine antibodies (44). Anti-TNFα monoclonal antibody (infliximab) was first administered to RA patients by intravenous infusion in an open label study initiated in May 1992 (45). The dose regime and schedule were derived by extrapolation from studies utilizing analogous anti-TNFα antibody reagents in the treatment of collagen induced arthritis (42). Twenty patients with chronic, erosive RA and evidence of active joint disease despite a history of treatment with multiple DMARDs (such as gold, methotrexate, and salazopyrin) received a total dose of 20 mg/kg infliximab by 2–4 intravenous infusions over a 2 wk period. In order to assess the effects of infliximab infusion without any introduction of bias by concomitantly administered DMARD or other antiinflammatory therapy, prior to administration of infliximab patients underwent a wash-out period of at least 4 wk, during which DMARDS were withdrawn and doses of nonsteroidal antiinflammatory drugs and/or corticosteroids were fixed. MOLECULAR BIOTECHNOLOGY

The primary aims of the open label study to evaluate safety and tolerability of infliximab in RA patients. Patients were found to be free of adverse effects during the time-frame of the trial and antibody infusions were well tolerated without alteration in cardiorespiratory function or body temperature. Clinical responses in the open label study were also encouraging. Patients showed marked and statistically significant improvement in all measure of clinical disease activity assessed. These included duration of early morning stiffness, tender and swollen joint count, and pain scores. The improvements in clinical status were accompanied by large and statistically significant falls in measures of acute phase response such as C reactive protein (CRP) and serum amyloid A. The magnitude and consistency of both clinical and serological measures strongly suggested that infliximab has a therapeutic effect. In order to confirm the impression of a clinical benefit to RA patients following administration of infliximab, a placebo controlled trial of short term administration of this antibody was conducted at 4 European centers with 73 participating patients (1). The trial design included randomization of patients to 1 of 3 treatment limbs, consisting of a single infusion of placebo (0.1% human serum Volume 19, 2001

158 albumin), or infliximab (at a dose or 1 or 10 mg/ kg) and standardized means of evaluating clinical response. In common with the open label trial, entry criteria to the first placebo controlled trial of infliximab in RA included the requirement for long-standing disease refractory to conventional therapy, and the need for a “wash-out” period for withdrawal of existing DMARD treatment and stabilization of doses of other concomitant antiinflammatory medications. The primary outcome measure of this study, assessed at wk 4, was the attainment of a standardized improvement in a composite index of disease activity incorporating 6 independent variables as defined by Paulus et al. (46). The Paulus criteria define a positive outcome as a significant improvement in at least 4 of 6 variables as follows: 1. At least 20% improvement in the 4 continuous variables tender and swollen joint count, duration of morning stiffness and ESR; 2. At least 2 grade improvement in the patient’s and observer’s assessment of disease severity.

At the predetermined 4 wk endpoint, clear differences in outcome, as judged by the Paulus 20% criteria, became evident in the 3 treatment limbs. Nineteen out of 24 patients (79%) in the high dose group and 11 out of 25 patients (44%) in the low dose group responded to therapy compared with only 2 out of 24 (8%) of patients treated with placebo. By arbitrarily raising the threshold of response to 50% improvement in the continuous variables comprising the Paulus index, it was found that 58% of patients benefited at the higher dose of infliximab and 28% at the lower dose. Given that the participating patients had previously received an average of 3 different DMARDs without adequate long-term disease suppression, these results seem even more impressive. Secondary outcome measures in the placebo controlled trial (1) included several individual clinical assessments of disease activity such as tender and swollen joint counts, pain and fatigue scores, patient and clinician assessment of overall disease activity, grip strength, and duration of early morning stiffness. A highly significant improvement was observed in each of these measures MOLECULAR BIOTECHNOLOGY

Taylor in recipients of both high and low dose anti-TNFα groups compared with recipients of placebo. Over the first 2 wk following infliximab infusion the proportion of patients responding as defined by the Paulus 20% criteria was broadly similar in recipients of either the low dose (1 mg/ kg) or high dose (10 mg/kg) regime. Similarly, the magnitude of improvement in various clinical parameters was comparable in the 2 groups at this early stage, generally 60–70% change from baseline. However, the 2 groups differed with respect to duration of response. A clear dose response relationship with time became apparent when participants in the blinded phase of the trial were followed to relapse of disease and further information concerning magnitude and duration of responses to an intermediate dose of infliximab (3 mg/kg) became available in an open label, follow-up phase of this trial. In this follow-up phase, recipients of placebo became eligible for treatments with one of several infliximab doses. Pooled analysis of the data from 14 recipients of 3 mg/kg infliximab, together with data from the recipients of placebo, 1 mg/kg or 10 mg/kg infliximab in the blinded phase of the study, permitted construction of dose-response curves over time (Fig. 4). The median duration of clinical response, assessed by the Paulus 20% criteria, was 3, 6, and 8 wk in the 1 mg/kg, 3 mg/kg, and 10 mg/kg infliximab dose regimes. In RA joint cell cultures the effect of TNFα are inhibited by concentrations of infliximab between 1 and 5 µg/mL. It is likely that duration of clinical response is directly related to maintenance of serum concentrations above approx 1 µg/mL. This would also explain why the proportion of responders and magnitude of responses are much the same at all 3 treatment doses in the first 2 wk after administration of infliximab. Following a single infusion of 1 and 10 mg/kg infliximab, concentrations (C max ) of 23 and 277 µg/mL are achieved. The terminal half-life of the chimerized antibody at the 10 mg/kg dose is 9–10 d, a time intermediate between the 1–2 d elimination half-life for most intact murine immunoglobulin and t h e 23 d half-life reported for human IgG1 gammaglobulin (47). Volume 19, 2001

Anti-TNF Therapy

Fig. 4. Duration of clinical response assessed by the Paulus disease activity index following placebo, n = 24 and cA2, anti-TNF mab at 1 mg/kg, n = 25, 3 mg/kg, n = 14 and 10 mg/kg, n = 24 (䉱). Based on data in (Maini, et al. Arthritis Rheum. 1995; 38, S186).

The clinical improvements observed in the first placebo controlled trial of anti-TNFα antibody treatment in RA were paralleled by improvements in various laboratory measures. There was significant improvement in the acute phase proteins CRP, serum amyloid A, fibrinogen, and haptoglobin (47). A dose dependent increase in hemoglobin levels was observed as compared to placebo accompanied by reductions in both erythropoietin and IL-6 levels. These observations implicate TNFα in the pathogenesis of anemia of chronic disease by a mechanism independent of erythropoietin suppression (48), perhaps acting directly on erythroid precursors in bone marrow. Thrombocytosis was also observed to be significantly reduced by infliximab in RA patients (1), possibly by downregulation of IL-6. The results of the first randomized double-blind study utilizing chimeric anti-TNFα mAbs to treat RA represents compelling evidence that selected cytokine blockade is of unequivocal therapeutic benefit in human chronic inflammatory disease. Nonetheless, despite impressive disease amelioration in the majority of patients following a single pulse of antibody treatment, the disease invariably relapsed. The important questions to ask, thereMOLECULAR BIOTECHNOLOGY

159 fore, is whether repeated antibody treatments result in long-term suppression of inflammation and in particular, modification of long-term disease outcomes. The first experience of repeated administration of infliximab was with a small cohort of patients originally enrolled in the open label trial. A total of 7 patients received between 2 to 4 complete treatment cycles as a single infusion of 10 mg/kg in cycles 2 to 4, following initial dose of 20 mg/ kg. After the initial treatment, the timing of the 10 mg/kg infusions was determined by the point of disease relapse, defined as loss of the Paulus 20% response measure consequent upon the proceeding treatment cycle (49). A reproducibly beneficial clinical response of similar magnitude was observed after each retreatment with anti-TNFα mAb suggesting that there is no TNFα independent pathway responsible for maintaining chronic joint inflammation. However, there was considerable variation in response duration between patients with a trend (although not statistically significant) toward shorter beneficial responses to successive antibody administrations. In the first treatment cycle, at a dose of 20mg/kg, the median response duration was 12 wk. By the fourth cycle, at a dose of 10 mg/kg, the median duration of benefit was 7.7 wk, a figure not statistically different from an 8 wk median response time reported following a single 10 mg/kg infusion in the placebo controlled trial. The murine variable region of the chimeric antiTNFα mAb used in the study might be expected to stimulate antimurine antibody responses upon repeated administration. Half the patients receiving retreatment cycles developed antimurine antibodies, mostly at low titer, as compared to only one recipient of infliximab participating in the original open-label trial (50). The feasibility of long-term suppression of inflammation in RA and a potential strategy for induction of immunological tolerance to infliximab were first investigated in a multicenter placebo controlled trial of repeated treatment cycles of infliximab, given either as monotherapy or combined with methotrexate (51). In this study, 5 infusions of placebo or infliximab at 1, 3, or 10 Volume 19, 2001

160 mg/kg were administered over a 14-wk period either with or without concomitant methotrexate therapy at a dose of 7.5 mg once weekly. Responses were sustained through to 26 wk in about 60% of patients. The frequency of human antichimeric antibody (HACA) responses was observed to be inversely related to the amount of antibody infused. Furthermore, the frequency of HACA responses was diminished in those patients receiving concomitant low dose oral methotrexate suggesting that methotrexate induces immunological tolerance to chimeric monoclonal antibodies. A phase III randomized, multicenter, placebocontrolled study (ATTRACT) was designed to further investigate feasibility of long-term treatment of RA with once weekly methotrexate and infliximab combination therapy (52). This study included patients with active disease despite relatively high dose methotrexate (median dose 15 mg/ wk). All 428 patients were maintained on MTX. Patients received either 3 or 10 mg/kg infliximab at 0, 2, 6 wk and then continued at either 4 or 8 wk intervals. A control group received MTX plus placebo infusions. The majority of patients achieved a significant response by 6 wk with further increments in the number of responders up to 1 yr. Over 60–70% change in individual parameters of disease activity was achieved in the infliximab groups as compared to placebo (52). This study confirmed that sustained reduction in symptoms and signs of RA can be achieved with methotrexate therapy together with repeated infliximab infusions. Although it is generally accepted that two years of treatment is required to demonstreate prevention of disability, analysis of the ATTRACT trial 54 wk data suggest that combined infliximab and methotrexate therapy increase the function of patients and result in a significantly better quality of life as compared to treatment with methotrexate as a monotherapy (53). Another 54-wk endpoint of the ATTRACT trial was joint protection as judged by a change in the rate of deterioration in radiographs of the hands and feet, assessed by a scoring system that assesses cartilage and bone loss separately. Joint space narrowing and erosions progressed as anticipated in recipients of methotrexate alone (placebo infusion). In contrast, therapy with infliximab plus MOLECULAR BIOTECHNOLOGY

Taylor methotrexate prevented the progressive joint damage characteristic of rheumatoid inflammation and resulted in improvement in the radiographic score from baseline in a significant percentage of patients (53). The combination therapy halted progression of joint damage both in patients with limited radiographic destruction at baseline and in those with extensive damage. Furthermore, prevention of progression of radiological damage over the 54-wk treatment period was observed not only in those patients exhibiting a clinical response, but also in those without a clinical response. These findings imply that TNFα has a critical role in the bone and cartilage damage associated with RA.

4.2. Other Antibody Treatments for RA Targeting TNFα A number of anti-TNFα biologic agents have now been shown to be effective in RA, confirming that TNFα is a good choice of therapeutic target in this disease. Other anti-TNFα mAbs include the humanized antibody CDP571 consisting of hypervariable regions from a mouse mAb, grafted onto a human IgG4 with κ light chains. CDP571 has a similar binding affinity to soluble TNFα (KD approx 100 picomolar) as does infliximab. Like infliximab, in a randomized placebo controlled trial (54), CDP571 was reported to be effective in reducing the early acute phase response in RA. The published data (54) would suggest that CDP571 is less effective than infliximab in amelioration of clinical disease assessment. A 10 mg/ kg IV infusion of CDP571 induced some improvement in swollen joint count but without achieving statistical significance. The magnitude of response was less than the change from baseline observed in similar patients receiving the same dose of infliximab (47). There is comparable downregulation of the acute phase protein CRP following administration of infliximab or CDP571 at the same dosage, as might be predicted given that IgG1 and IgG4 isotypes neutralize soluble antigen with similar efficiency. Once possible explanation for the observed differences in clinical efficacy of infliximab and CDP571 is that isotype dependent cell lysis, mediated by IgG1, may contribute to amelioration of joint inflammation. Volume 19, 2001

Anti-TNF Therapy A different approach for targeting TNFα in inflammatory disease would be to infuse exogenous soluble TNF receptor proteins (sTNFR). To increase the half-life, affinity and biovaliability of the sTNFR, DNA encoding the soluble portion of human TNFRp75 was linked to DNA encoding the Fc portion of human IgG1 and the combined DNA was then expressed in a mammalian cell line to form a recombinant fusion protein (sTNFR:Fc). This protein, an immunoglobulin like dimer consists entirely of human amino acid sequences and acts as a competitive inhibitor of TNFα. Incontrast to mAb with specificity for TNFα, sTNFR:Fc can also bind lymphotoxinα (TNFβ). In a multicenter, dose ranging phase II doubleblind trial, 180 patients with refractory RA were randomly assigned to receive twice weekly subcutaneous injections of placebo, or 1 of 3 doses of TNFR:Fc for a 3 mo time period (55). The 3 treatment groups comprised doses of 0.25, 2 or 16 mg/ m2 body surface area. Clinical responses were measured by change in a composite index of disease activity (similar to the Paulus criteria discussed previously) as defined by the American College of Rheumatology (ACR) criteria. Administration of sTNFR:Fc induced a dose dependent reduction in disease activity. Responses at the 20% ACR level were observed in 75% of patients in the high dose group at 3 mo, a highly significant finding compared with a 14% response in the placebo group. The mean percentage reduction in the number of tender or swollen joints at 3 mo was 61% as compared with 25% in the placebo group. As reported in the case of anti-TNFα mAb therapy, sTNR:Fc treatment was associated with significant reduction in measures of acute phase response such as erythrocyte sedimentation rate (ESR) and CRP. Cessation of therapy was associated with an increase of disease activity toward the pretrial base line. No dose limiting toxic effects were observed in this study, and no antibodies to TNF:Fc were detected in serum samples. Adverse events thought to be related, or potentially related, to administration of the fusion protein included mild injection site reactions and upper respiratory tract symptoms. A subsequent placebo-controlled study confirmed the efficacy of etanercept at 3 and 6 mo MOLECULAR BIOTECHNOLOGY

161 when administered at a dose of 25 mg by subcutaneous injection twice per week (56). As in the case of infliximab, it has recently been reported that treatment of established RA with the combination of etanercept and methotrexate is safe, well tolerated, and provides significantly greater clinical benefit than methotrexate alone (57). In this study, patients were treated with either etanercept or placebo while continuing to take methotrexate. There was no comparison of combination therapy with etanercept alone and thus it is not yet clear whether the enhanced clinical response to combined treatment represents an additive, or synergistic benefit. The efficacy and safety of etanercept and methotrexate have been compared in patients with early RA (58). As compared with recipients of methotrexate monotherapy, patients treated with 25 mg etanercept twice weekly by subcutaneous injection exhibited a more rapid rate of improvement with significantly more patients achieving ACR 20, 50, and 70% responses during the first 6 mo of therapy. In the patient cohort enrolled in this study, the predicted rate of radiological progression would be 4–5 points on the Sharp erosion subscale and 4 points on the joint space narrowing subscale. The mean increase in erosion score during the first 6 mo of treatment was 0.3 in the patients treated with 25 mg etanercept twice weekly and 0.68 in the methotrexate monotherapy group (p = 0.001). Over a 12 mo treatment period, the respective figures were 0.47 and 1.03 (p = 0.002). Patients treated with 25 mg etanercept twice weekly had significantly fewer adverse reactions and significantly fewer infections than the group treated with methotrexate alone.

4.3. Antibodies to TNFα in the Treatment of Other Inflammatory Conditions 4.3.1. Sepsis The largest clinical trial of anti-TNFα mAb for the treatment of sepsis included 994 patients. In a retrospective analysis of a subgroup considered to have had septic shock, there was a 45% reduction in mortality 3 d after the antibody (59). However, there was no significant reduction in mortality in the antibody treated cohort of the 28 d primary Volume 19, 2001

162 end-point of this study. The findings of this large study are consistent with a number of others using anti-TNFα mAbs or TNFR-IgG fusion protein constructs (47) in which little or no clinical benefit was observed. In the case of a murine antiTNFα mAb (CD006: Celltech), only a subgroup of recipients with elevated circulating TNFα appeared to benefit from antibody infusion, but there was no overall improvement in survival (60). A recently reported phase II study suggests a possible advance in characterizing a subgroup of patients for sepsis syndrome who may potentially benefit from TNFα (61) blockade. In this study, 122 patients were randomized to receive different doses of anti-TNFα antibody fragment (MAK195F) or placebo. As in earlier trial, no overall increase in survival was observed in recipients of the antibody fragment as compared to placebo. Retrospective reexamination of the data following the stratification of patients according to plasma IL-6 concentrations suggested a dose-dependent benefit of the antibody fragments in those patients with IL-6 concentrations, exceeding 1000 picograms/mL. Since TNFα induces IL-6 production identification of patients with this level of plasma IL-6 may select a group with high cumulative TNFα exposure. Follow-up trials are underway to further investigate this particular subgroup.

4.3.2. Jarisch-Herxheimer Reaction Antibiotic treatment of patients with louse borne relapsing fever is often complicated by acute fever, rigors, and hypotension, a syndrome termed the Jarisch-Herxheimer reaction. This syndrome is associated with elevated plasma levels of TNFα, IL-6, and IL-8 (29). Anti-TNFα pretreatment has been shown to markedly attenuate this sepsis syndrome-like condition in patients with louse borne relapsing fever (62). In a doubleblind, placebo-controlled trial, the frequency of reaction was reduced to 50% in patients pretreated with sheep anti-TNFα Fab as compared to 90% of patients receiving placebo prior to antibiotic therapy. 4.3.3. Inflammatory Bowel Disease Crohn’s disease is characterized by the classic triad of diarrhea, abdominal pain, and weight loss. MOLECULAR BIOTECHNOLOGY

Taylor TNFα is believed to be an important mediator of inflammatory lesions which can occur in any portion of the alimentary tract. In support of this hypothesis, rapid and sustained improvements following anti-TNFα mAb therapy were first observed in patients with steroid-dependent Crohn’s disease in open label studies involving single infusion of between 1 and 20 mg/kg infliximab (63,64). These observations were confirmed in a double-blind, placebo-controlled trial in which 108 patients with active Crohn’s disease were randomized to receive placebo of 3 different doses of infliximab (5, 10, or 20 mg/kg) as a single intravenous (65). Of the placebo treated patients a degree of clinical response was observed in 17% and 4% achieved remission. By contrast, 65% of the anti-TNFα treated groups achieved clinical responses. The response was sustained for 8 wk in 80% of patients and beyond 12 wk in 63%. Similar therapeutic benefits have been reported at 2 wk in phase III trials with humanized antiTNFα CDP571. However, there was no beneficial effect in the treated groups at later time-points (66).

4.4. Mechanisms of Action of Anti-TNFα Antibodies in the Treatment of Rheumatoid Arthritis The availability of agents that block biological activity of TNFα in rheumatoid arthritis has permitted studies that confirm the key role of this cytokine in the pathogenesis of disease. Evidence to date support the hypothesis that there are 2 particularly important mechanisms of action; deactivation of the proinflammatory cytokine cascade at the site of inflammation and diminished recruitment of inflammatory cells from blood to the rheumatoid joint. The first indirect evidence to suggest that TNFα blockade in vivo has an effect on the cytokine network was the marked reduction in acute phase proteins, such as CRP, within days of administration of anti-TNFα antibody. Elevated serum concentrations of CRP in RA reflect production by hepatocytes under the regulation of cytokines signalling via the gp130 receptor, in particular, IL-6. Since circulating IL-6 is detectable in the majority of patients with active RA, it has been possible to directly confirm that changes Volume 19, 2001

Anti-TNF Therapy in CRP are accompanied by changes in IL-6 (45,67), maximum reductions occurring within 24 h of infliximab infusion. In contrast, IL-1 is either undetectable or present in very low concentrations in peripheral blood of most patients. Therefore, on the basis of serum analysis, it is not possible to directly test the prediction, based on the in vitro studies discussed earlier (10), that TNFα blockade regulates IL-1 production. Instead, we addressed this question using synovial tissue biopsies obtained before and 2 wk after a single infusion of 10 mg/kg infliximab. A modified immunohistochemical method detecting cytokineproducing rather than cytokine-binding cells was applied to determine synthesis of TNFα, IL-1α, and IL-1β in fixed, cryopreserved sections. Assessment of these sections by computerized image analysis demonstrated that synovial TNFα synthesis was reduced, but not abrogated, after infliximab treatment in all of a group of 8 patients (68). There was significant correlation between baseline TNFα expression and change in TNFα in response to therapy. Furthermore, those patients with highest baseline TNFα synthesis achieved the best clinical responses. Reduction in synovial IL-1α and IL-1β was observed in a subgroup of these patients. Although these data support the contention that TNFα differentially regulates IL-1α and IL-1β, the small sample size and a single follow-up biopsy 2 wk after therapy does not permit a definitive conclusion regarding the degree of potential regulation of IL-1 by TNFα. There is clear evidence that the therapeutic benefit of anti-TNF antibody treatment is not mediated by an upregulation of endogenous proinflammatory cytokine inhibitors, since circulating IL-1ra and soluble TNF receptor levels fall after infliximab infusion (47). During early clinical trials of anti-TNFα therapy for RA, two observations suggested that this treatment modulates inflammatory cell traffic to joints. First, peripheral blood lymphocyte counts were observed to rise to a maximum level within 24 h of infliximab infusion (69). Secondly, histological examination of multiple synovial biopsies obtained during arthroscopic examination of knee joints of RA patients immediately prior to and 4 MOLECULAR BIOTECHNOLOGY

163 wk after a single infusion of 10 mg/kg infliximab revealed significant reduction in synovial cellularity after treatment. In particular, there was significant reduction in synovial CD3 cells (70,71). Further analysis of synovial biopsies taken before and 2 wk after a single infusion of 10 mg/ kg infliximab confirmed that there is also a reduction in CD22 B cells and macrophage infiltration in both synovial lining and sublining layers (72). Given that these cell types are relatively long-lived, these findings may underestimate the maximum effect of TNFα blockade on synovial cellularity. Histological assessment of synovial biopsies obtained before and after administration of infliximab demonstrate a significant reduction in expression of the chemokines IL-8 and MCP-1 (72). There is also a significant reduction in synovial tissue expression of endothelial E-selectin and VCAM-1 with a down trend in ICAM-1 expression in both the lining and sublining layers (70,71). Soluble forms of the adhesion molecules ICAM1, VCAM-1, and E-selectin are derived from the corresponding cell surface forms by proteolytic cleavage (73). The release of shed adhesion molecules correlates with changes in expression of the corresponding cell surface molecules on cultured endothelial cells following activation with TNFα or IL-1 (74). Serum samples from RA patients participating in the first placebo controlled trial of infliximab (1) were assayed by ELISA for all 3 soluble adhesion molecules. Median baseline concentrations were markedly elevated, and significant dose-dependent reduction in soluble Eselectin and ICAM-1 were observed 4 wk after infliximab administration (69). The magnitude of change was generally greatest in those patients assessed as having a good clinical response. Levels of soluble VCAM-1 were unchanged. Downregulation of chemokines and endothelial adhesion molecule expression following TNFα blockade would be predicted to reduce margination and migration of all leukocyte classes to sites of inflammation, with a consequent increase in circulating cell counts. However, in contrast, to peripheral blood lymphocyte counts, numbers of peripheral blood granulocytes, which have a cirVolume 19, 2001

164 culating half-life measured in hours, decrease after infliximab with maximal changes within 24 h. Nevertheless, by studying the kinetics of autologous granulocytes, separated and labeled with 111In, we have shown that there is reduction in the marginating granulocyte pool after a single 10 mg/kg infusion of infliximab (75). These observations are likely to reflect two different processes; diminished granulocyte margination (with consequent increase in cirulating granulocyte counts) and reduced myeloid cell production secondary to downregulation of GM-CSF (with a consequent reduction in cirulating granulocyte counts) since TNFα induces GM-CSF production by cultured RA synovial cells (68,69). Because of the short circulating half-life of the granulocyte, it seems likely that a diminished rate of cell production will dominate the peripheral blood picture at the time point sampled. Analysis of images of knee, hand, and wrist joints after bolus injection of autologous 111In granulocytes confirm that TNFα blockade results in significantly reduced granulocyte traffic to inflamed joints. Thus TNFα blockade reduces both endothelial adhesiveness and chemotactic gradients with corresponding reduction in inflammatory cell margination and migration in to RA joints.

4.5. Potential Limitations of Anti-TNFα Antibody Therapy in RA 4.5.1. General One potential limitation of the anti-TNFα antibody agents discussed is the need for parentral administration. Alternative means of targeting TNFα include the use of small molecules, effective by oral administration, inhibiting the proteases that process the membrane bound cytokine or the protein kinases regulating TNFα synthesis. 4.5.2. Safety Issues Given that TNFα evolved to serve specific biological functions, the near elimination of this factor by specific antibody treatment raised the concern that any benefits of therapy might be accompanied by side effects related to defective im-

MOLECULAR BIOTECHNOLOGY

Taylor mune surveillance. In fact, neither opportunistic infections nor increased neoplasia have been observed and there is little evidence of defective responses to infectious diseases in recipients of infliximab (reviewed in detail in Feldmann [47]). Where infections have been observed in patients with Crohns or RA treated with anti-TNFα antibody they have been typical of the infections commonly occurring in these disease groups. In a placebo controlled trial in which infliximab was administered to patients with presumed sepsis, there was a trend towards mortality benefit at 4 wk in recipients of infliximab, no opportunistic infections were observed and adverse events were similar in the anti-TNFα and placebo treated groups (47). An unexpected side effect reported in RA trials of several anti-TNFα biologics was the development of antibodies to double-stranded DNA (antids-DNA). IgM class anti-ds-DNA antibodies were observed several wk after infliximab infusion in 7 of 69 (6%) patients treated in early trials (1,45,49). Three of these individuals received a subsequent infliximab infusion. In 1, the anti-ds-DNA became undetectable after the second infliximab infusion; in a second, the titer remained unaltered; and in the third, titers rose after the second infliximab infusion. None of these patients developed any signs or symptoms of a lupus-like syndrome. In a multidose study of infliximab in RA, 1 patient developed worsening arthritis, dyspnoea, and pleuritic chest pain 4 wk after a 4th dose of 3 mg/ kg infliximab. These symptoms abated and a further 3 mg/kg infusion was given, followed 3 wk later by more severe and persistent symptoms. Serial serum analysis revealed that ANA and antibodies to ds-DNA appeared for the first time 6 wk into the study, suggesting that infliximab administration may have precipitated a lupus-like syndrome in this individual (47). The symptoms resolved with oral steroid treatment. The pathogenesis of induction of anti-ds-DNA antibodies in a minority of RA patients treated with anti-TNFα therapy remains unknown although other antirheumatic drugs, such as sulphasalazine, can also induce such autoantibodies.

Volume 19, 2001

Anti-TNF Therapy 5. Concluding Remarks The development of anticytokine therapy represents an important advance in both treatment and understanding of basic disease mechanisms in chronic inflammation. In the example of RA, TNFα is the most extensively studied target to date, but others such as IL-1 and IL-6 are also under evaluation. Alternatives to antibody treatment include the use of naturally occurring antiinflammatory cytokines, for example, IL-4, IL-10, and IL-11. Given the now compelling evidence of proinflammatory cytokine involvement in the pathogenesis of RA, and the marked clinical benefit of TNF blockade in established disease, it may well be that enhanced long-term benefit will be derived by anticytokine intervention earlier in the evolution of disease. In the case of RA, short courses of anti-TNF treatment can be used to induce remission, or to control disease flares, so allowing conventional medication to exert its beneficial effect. Long-term monotherapy with anti-TNF agents or combination therapy using both conventional drugs and anticytokine agents leads to sustained improvement in symptoms and signs of RA in a majority of patients. Furthermore, such therapies have the potential to significantly retard radiological progression of disease with a concomitant improvement in function and quality of life. References 1. Elliott, M. J., Maini, R. N., Feldmann, M., et al. (1994) Randomised double-blind comparison of chimeric monoclonal antibody to tumor necrosis factor α (cA2) versus placebo in rheumatoid arthritis. Lancet 344, 1105–1110. 2. Pincus, T. and Callahan, L. F. (1993) What is the natural history of rheumatoid arthritis? Rheum. Dis. Clin. North Am. 19, 123–151. 3. Sharp, J. T., Wolfe, F., Mitchell, D. M., and Bloch, D. A. (1991) The progression of erosion and joint space narrowing scores in rheumatoid arthritis during the first twentyfive years of disease. Arthritis Rheum. 34, 660–668. 4. Janossy, G., Panayi, G., Duke, O., Bofill, M., Poulter, L. W., and Goldstein, G. (1981) Rheumatoid arthritis: a disease of T-lymphocyte/macrophage immunoregulation. Lancet ii, 839–841. 5. Cush, J. J. and Lipsky, P. E. (1988) Phenotypic analysis of synovial tissue and peripheral blood lymphocytes isolated from patients with rheumatoid arthritis. Arthritis Rheum. 31, 1230–1238.

MOLECULAR BIOTECHNOLOGY

165 6. Firestein, G. S. and Zvaifler, N. J. (1990) How important are T cells in chronic rheumatoid synovitis? Arthritis Rheum. 33, 768–773. 7. Miossec, P., Naviliat, M., D’Angeac, A. D., Sany, J., and Banchereau, J. (1990) Low levels of interleukin-4 and high levels of transforming growth factor β in rheumatoid synovitis. Arthritis Rheum. 33, 1180–1187. 8. Feldmann, M., Brennan, F. M., and Maini, R. N. (1996) Role of cytokines in rheumatoid arthritis. Ann. Rev. Immunol. 14, 397–440. 9. Buchan, G., Barrett, K., Turner, M., Chantry, D., Maini, R. N., and Feldmann, M. (1988) Interleukin-1 and tumor necrosis factor mRNA expression in rheumatoid arthritis: prolonged production of IL-1α. Clin. Exp. Immunol. 73, 449–455. 10. Brennan, F. M., Chantry, D., Jackson, A., Maini, R., and Feldmann, M. (1989) Inhibitory effect of TNFα antibodies on synovial cell interleukin-1 production in rheumatoid arthritis. Lancet 2, 244–247. 11. Butler, D., Maini, R. N., Feldmann, M., and Brennan, F. M. (1995) Blockade of TNFα with chimeric anti TNFα monoclonal antibody, cA2 reduces (IL-6 and IL8) release in RA NMC cultures: A comparison with IL-1ra. submitted. Eur. Cytokine. Netw. 6, 225–230. 12. Dayer, J. M., Beutler, B., and Cerami, A. (1985) Cachectin/tumor necrosis factor stimulates collagenase and prostaglandin E2 production by human synovial cells and dermal fibroblasts. J. Exp. Med. 162, 2163– 2168. 13. Katsikis, P. D., Chu, C. Q., Brennan, F. M., Maini, R. N., and Feldmann, M. (1994) Immunoregulatory role of interleukin 10 in rheumatoid arthritis. J. Exp. Med. 179, 1517–1527. 14. Fava, R., Olsen, N., Keski-Oja, J., Moses, H., and Pincus, T. (1989) Active and latent forms of transforming growth factor β activity in synovial effusions. J. Exp. Med. 169, 291–296. 15. Wahl, S. M. (1994) Transforming growth factor β: The Good, the Bad, and the Ugly. J. Exp. Med. 180, 1587– 1590. 16. Cope, A. P., Aderka, D., Doherty, M., et al. (1992) Increased levels of soluble tumor necrosis factor receptors in the sera and synovial fluid of patients with rheumatoid diseases. Arthritis Rheum. 35, 1160–1169. 17. Coley, W. (1893) The treatment of malignant tumors by repeated inoculations of erysipelas: with a report of ten original cases. Am. J. Med. Sci. 105, 487–511. 18. Aggarwal, B. B., Kohr, W. J., Hass, P. E., et al. (1985) Human tumor necrosis factor. Production, purification, and characterization. J. Biol. Chem. 260, 2345–2354. 19. Pennica, D., Nedwin, G. E., Hayflick, J. S., P. H. S., et al. (1984) Human tumor necrosis factor: precursor structure expression and homology to lymphotoxin. Nature 312, 724–729. 20. Shirai, T., Yamaguchi, H., Ito, H., Todd, C. W., and Wallace, R. B. (1985) Cloning and expression in Esche-

Volume 19, 2001

166

21.

22.

23. 24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

richia coli of the gene for human tumor necrosis factor. Nature 313, 803–809. Wang, A. M., Creasey, A. A., Ladner, M. B., et al. (1985) Molecular cloning of the complementary DNA for human tumor necrosis factor. Science 228, 149–154. Beutler, B. and Cerami, A. (1986) Cachectin and tumor necrosis factor as two sides of the same biological coin [review]. Nature 320, 584–588. Jones, E. Y., Stuart, D. I., and Walker, N. P. C. (1989) Structure of tumor necrosis factor. Nature 338, 225–228. Bazzoni, F. and Beutler, B. (1996) The tumor necrosis factor ligand and receptor families (review). N. Engl. J. Med. 334, 1717–1725. Engelmann, H., Aderka, D., Rubinstein, M., Rotman, D., and Wallach, D. (1989) A tumor necrosis factorbinding protein purified to homogeneity from human urine protects cells from tumor necrosis factor toxicity. J. Biol. Chem. 264, 11,974–11,980. Seckinger, P., Isaaz, S., and Dayer, J. M. (1988) A human inhibitor of tumor necrosis factor α. J. Exp. Med. 167, 1511–1516. Poltorak, A., Peppel, K., and Beutler, B. (1994) Receptor-mediated label-transfer assay (RELAY): a novel method for the detection of plasma tumor necrosis factor at attomolar concentrations. J. Immunol. Methods 169, 93–99. Tracey, K. J. and Cerami, A. (1994) Tumor necrosis factor: a pleiotropic cytokine and therapeutic target (review). Ann. Rev. Med. 45, 491–503. Negussie, Y., Remick, D. G., DeForge, L. E., Kunkel, S. L., Eynon, A., and Griffin, G. E. (1992) Detection of plasma tumor necrosis factor, interleukins 6, and 8 during the Jarisch-Herxheimer reaction of relapsing fever. J. Exp. Med. 175, 1207–1212. Chatenoud, L., Ferran, C., Reuter, A., et al. (1989) Systemic reaction to the anti-T-cell monoclonal antibody OKT3 in relation to serum levels of tumor necrosis factor and interferon-gamma. N. Engl. J. Med. 320, 1420,1421. Dayer, J. M. and Burger, D. (1994) Interleukin-1, tumor necrosis facto and their specific inhibitors [review]. Eur. Cytokine Netw. 5, 563–571. Marchant, A., Bruyns, C., Vanderabeele, P., et al. (1994) Interleukin-10 controls interferon-gamma and tumor necrosis factor production during experimental endotoxemia. Eur J. Immunol. 24, 1167–1171. Espevik, T., Figari, I. S., Shalaby, M. R., et al. (1987) Inhibition of cytokine production by cyclosporin A and transforming growth factor beta. J. Exp. Med. 166, 571–576. Han, J., Thompson, P., and Beutler, B. (1990) Dexamethasone and pentoxifylline inhibit endotoxin induced TNF synthesis at separate points in their signaling pathway. J. Exp. Med. 172, 391–394. Parmely, M. J., Zhou, W. W., Edwards, C. K., Borcherding, D. R., Silverstein, R., and Morrison, D. C.

MOLECULAR BIOTECHNOLOGY

Taylor

36.

37.

38.

39.

40.

41.

42.

43.

44. 45.

46.

47.

48.

(1993) Adenosine and a related cabocyclic nucleoside analogue selectively inhibit tumor necrosis factoralpha production and protect mice against endotoxin challenge. J. Immunol. 151, 389–396. Bouma, M. G., Stad, R. K., van den Wildenberg, F. A., and Buurman, W. A. (1994) Differential regulatory efffects of adenosine on cytokine release by activated human monocytes. J. Immunol. 153, 4159–4168. Vannier, E., Miller, L. C., and Dinarello, C. A. (1991) Histamine suppressess gene expression and synthesis of tumor necrosis factor alpha via histamine H2 receptors. J. Exp. Med. 174, 281–284. Florquin, S., Amraoui, Z., Dubois, C., Decuyper, J., and Goldman, M. (1994) The protective role of endogenously synthesized nitric oxide in staphylococcal enterotoxin B-induced shock in mice. J. Exp. Med. 180, 1153–1158. Eigler, A., Moeller, J., and Endres, S. (1995) Exogenous and endogenous nitric oxide attenuate tumor necrosis factor synthesis in the murine macrophage cell line RAW 264.7. J. Immunol. 154, 4048–4054. Moreira, A. L., Sampaio, E. P., Zmuidzinas, A., Frindt, P., Smith, K. A., and Kaplan, G. (1993) Thalidomide exerts its inhibitory action on tumor necrosis factor alpha by enhancing mRNA degradation. J. Exp. Med. 177, 1675–1680. Keffer, J., Probert, L., Cazlaris, H., et al. (1991) Transgenic mice expressing human tumor necrosis factor: a predictive genetic model of arthritis. EMBO J. 10, 4025–4031. Williams, R. O., Feldmann, M., and Maini, R. N. (1992) Anti-TNF ameliorates joint disease in murine collagen-induced arthritis. Proc. Natl. Acad. Sci. USA 89, 9784–9788. Knight, D. M., Trinh, H., Le, J., et al. (1993) Construction and initial characterization of a mousehuman chimeric anti-TNF antibody. Mol. Immunol. 30, 1443–1453. Winter, G. and Harris, W. J. (1993) Humanized antibodies. Immunol. Today 14, 243–246. Elliott, M. J., Maini, R. N., Feldmann, M., et al. (1993) Treatment of rheumatoid arthritis with chimeric monoclonal antibodies to tumor necrosis factor α. Arthritis Rheum. 36, 1681–1690. Paulus, H. E., Egger, M. J., Ward, J. R., and Williams, H. J. (1990) The Cooperative Systemic Studies of Rheumatic Diseases Group. Analysis of improvement in individual rheumatoid arthritis patients treated with disease-modifying antirheumatic drugs, based on the findings in patients treated with placebo. Arthritis Rheum. 33, 477–484. Feldmann, M., Elliott, M. J., Woody, J. N., and Maini, R. N. (1997) Anti-tumor necrosis factor-a therapy in rheumatoid arthritis. Adv. Immunol. 64, 283–350. Davis, D., Charles, P. J., Potter, A., Feldmann, M., Maini, R. N., and Elliott, M. J. (1997) Anaemia of

Volume 19, 2001

Anti-TNF Therapy

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

chronic disease in rheumatoid arthritis: in vivo effects of tumor necrosis factor α. Br. J. Rheum. 36, 950–956. Elliott, M. J., Maini, R. N., Feldmann, M., et al. (1994) Repeated therapy with monoclonal antibody to tumor necrosis factor a (cA2) in patients with rheumatoid arthritis. Lancet 344, 1125–1128. Maini, R. N., Elliott, M. J., Brennan, F. M., et al. (1995) Monoclonal anti-TNFα antibody as a probe of pathogenesis and therapy of rheumatoid disease. Immunol. Revs. 144, 195–223. Maini, R. N., Breedveld, F. C., Kalden, J. R., Smolen, J. S., Davis, D., et al. (1998) Therapeutic efficacy of multiple intravenous infusions of anti-tumor necrosis factor a monoclonal antibody combined with lowdose weekly methotrexate in rheumatoid arthritis (RA). Arthritis Rheum. 41, 1552–1563 Maini, R. N., St. Clair, E. William, Breedveld, F., et al. (1999) For the ATTRACT Study Group Infliximab (chimeric anti-tumor necrosis factor a monoclonal antibody) versus placebo in rheumatoid arthritis patients rerceiving concomitant methotrexate: a randomised phase III trial. Lancet 354, 1932–1939. Lipsky, P. E., van der Heidje, D. M. F. M., St. Clair, E. William, et al. (2000) For the ATTRACT Study Group Infliximab and methotrexate in the treatment of Rheumatoid Arthritis. N. Engl. .J Med. 343, 1594–1602. Rankin, E. C. C., Choy, E. H. S., Kassimos, D., et al. (1995) The therapeutic effects of an engineered human anti-tumor necrosis factor alpha antibody (CDP571) in rheumatoid arthritis. Br. J. Rheumatol. 34, 334–342. Moreland, L. W., Baumgartner, S. W., Schiff, M. H., et al. (1997) Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. N. Engl. J. Med. 337, 141–147. Moreland, L. W., Schiff, M. H., Baumgartner, S. W., et al. (1999) Etanercept therapy in rheumatoid arthritis. A randomized, controlled trial. Ann. Intern. Med. 130, 478–486. Weinblatt, M. E., Kremer, J. M., Bankhurst, A. D., et al. (1999) A trial of etanercept, a recombinant tumor necrosis factor receptor:Fc fusion protein, in patients with rheumatoid arthritis receiving methotrexate. N. Eng. J. Med. 340, 253–259. Bathon, J. M., Martin, R. W., Fleischmann, R. M., et al. (2000) A comparison of etanercept and methotrexate in patients with early Rheumatoid Arthritis. N. Engl. J. Med. 343, 1586–1593. Abraham, E., Wunderink, R., Silverman, H., et al. (1995) Efficacy and safety of monoclonal antibody to human tumor necrosis factor alpha in patients with sepsis sydrome. A randomized, controlled, doubleblind, multicenter, clinical trial. TNF-alpha MAb Sepsis Study Group. JAMA 273, 934–941. Fischer, Jr. C. J., Opal, S. M., Dhainaut, J.-F., et al. (1993) Influence of an anti-tumor necrosis factor

MOLECULAR BIOTECHNOLOGY

167

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

monoclonal antibody on cytokine levels in patients with sepsis. Crit. Care Med. 21, 318–327. Reinhart, K., Wiegand-Lohnert, C., Grimminger, F., et al. (1996) Assessment of the safety and efficacy of the antibody-fragment, MAK 195F, in patients with sepsis and septic shock: a multicenter, randomized, placebo-controlled, dose-ranging study. Crit. Care Med. 24, 733–742. Fekade, D., Knox, K., Hussein, K., et al. (1996) Prevention of Jarisch-Herxheimer reactions by treatment with antibodies against tumor necrosis factor alpha. N. Engl. J. Med. 335, 311–315. Van Dullemen, H. M., Van Deventer, S. J. H., Hommes, D. W., et al. (1995) Treatment of Crohn’s disease with anti-tumor necrosis factor chimeric monoclonal antibody (cA2). Gastroenterology 109, 129–135. McCabe, R. P., Woody, J. N., van Deventer, S., et al. (1996) A multicenter trial of cA2 anti-TNF chimeric monoclonal antibody in patients with active Crohn’s disease. Gastroenterology Suppl. 110, A962. Targan, S. R., Hanauer, S. B., van Deventer, S. J., et al. (1997) A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn’s disease. Crohn’s Disease cA2 Study Disease. N. Engl. J. Med. 337, 1029–1035. Stack, W. A., Mann, S. D., Roy, A. J., et al. (1997) Randomised controlled trial of CDP571 antibody to tumor necrosis factor-alpha in Crohn’s disease. Lancet 349, 521–524. Charles, P., Elliott, M. J., Davis, D., et al. (1999) Regulation of cytokines, cytokine inhibitors, and acutephase proteins following anti-TNF-alpha therapy in rheumatoid arthritis. J. Immunol. 163, 1521–1528. Ulfgren, A.-K., Andersson, U., Engström, M., Klareskog, L., Maini, R. N., and Taylor, P. C. (2000) Systemic anti-TNFα therapy in rheumatoid arthritis down-regulates synovial TNFα synthesis. Arthritis Rheum. 43, 2391–2396. Paleolog, E. M., Elliott, M. J., Woody, J. N., Feldmann, M., and Maini, R. N. (1996) Deactivation of vascular endothelium by monoclonal anti-tumor necrosis factor α antibody in rheumatoid arthritis. Arthritis Rheum. 39, 1082–1091. Tak, P. P., Taylor, P. C., Breedveld, F. C., et al. (1996) Decrease in cellularity and expression of adhesion molecules by anti-tumor necrosis factor α monoclonal antibody treatment in patients with rheumatoid arthritis. Arthritis Rheum. 39, 1077–1081. Taylor, P. C., Maini, R. N., Tak, P. P., et al. (1997) Reply to Decrease in cell adhesion molecules by treatment with anti-tumor necrosis factor a monoclonal antibody. Arthritis Rheum. 4, 789,790. Taylor, P. C., Peters, A. M., Paleolog, E., et al. (2000) TNFα blockade in patients with rheumatoid arthritis reduces chemokines and leukocyte traffic to joints. Arthritis Rheum. 43, 38–47.

Volume 19, 2001

168 73. Gearing, A. J. H. and Newman, W. (1993) Circulating adhesion molecules in disease. Immunol. Today 14, 506–512. 74. Leeuwenberg, J. F. M., Smeets, E. F., Neefjes, J. J., et al. (1992) E-selectin and intercellular adhesion molecule-1 are released by activated human endothelial cells in vitro. Immunology 77, 543–549. 75. Taylor, P. C., Peters, A. M., Glass, D. M., and Maini, R. N. (1999) Effect of treatment of Rheumatoid patients with anti-TNFα on reticuloendot-

MOLECULAR BIOTECHNOLOGY

Taylor helial and intrapulmonary granulocyte traffic. Clin. Sci. 97, 85–89. 76. Alvaro-Gracia, J. M., Zvaifler, N. J., Brown, C. B., Kaushansky, K., and Firestein, G. S. (1991) Cytokines in chronic inflammatory arthritis. J. Immunol. 146, 3365–3371. 77. Haworth, C., Brennan, F. M., Chantry, D., Turner, M., Maini, R. N., and Feldmann, M. (1991) Expression of granulocytemacrophage colony-stimulating factor in rheumatoid arthritis: regulation by tumor necrosis factor-α. Eur. J. Immunol. 21, 2575–2579.

Volume 19, 2001