• biologicals;
  • cytokines;
  • immunotherapy;
  • inflammatory skin diseases;
  • psoriasis;
  • TNF-α


  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

Abstract:  Numerous recent investigations have pointed to a key role of the proinflammatory, pleiotropic cytokine tumor necrosis factor-α (TNF-α) in host defense and inflammatory processes. TNF overexpression has been found in lesional skin and in the circulation of psoriatic patients, and it was suggested that TNF-α is crucial in this and other immune diseases. Several approaches to inhibit TNF-α activity have been developed. These include three different neutralizing antibodies to TNF-α as well as three different soluble TNF-α receptors with characteristic properties designed to bind the 17-KDa soluble trimeric TNF-α and the 26-KDa membrane-bound form of TNF-α. Clinical trials have demonstrated significant antipsoriatic effects, and it is likely that blocking TNF-α will become an important therapeutic option. The data available from these trials contribute to further understanding of the disease by demonstrating the major role of TNF-α. An in-depth understanding of the regulation of TNF gene expression, protein production, receptor expression, and signaling pathways may lead to further, potentially important novel therapeutic strategies and antipsoriatic active small molecules, suitable for oral application in the future. Here we review the current knowledge of TNF biology, the approaches to inhibit TNF activity, and their clinical and immunological effects in psoriasis. In addition, the host-defense effects and chronic TNF-blocking activity are discussed.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

Tremendous progress in the understanding of the pathophysiology of psoriasis has been achieved during the last few decades. Psoriasis is now known to be a T-cell-mediated (auto) immune disease in which cytokines play an essential part (1–5). Arguments for considering psoriasis as a T-cell-mediated dermatosis include:

  • • 
    Presence of activated T cells in the skin lesions
  • • 
    Cure of the disease by bone marrow transplantation from healthy persons and transfer of the disease by transplantation of bone marrow from psoriatic patients
  • • 
    Demonstration of the impact of immunocytes by SCID mice experiments
  • • 
    Therapeutic effects of immunosuppressants targeting T lymphocytes (e.g. cyclosporin-A, anti-T-cell antibodies).

This new pathophysiological understanding offers for the first time a chance of developing new and better-directed therapeutic strategies (4). One recent successful approach has been the neutralization of the major proinflammatory cytokine, tumor necrosis factor-α (TNF-α). The first clinical data are very impressive. They highlight the tremendous role of this cytokine in the pathogenesis of disease. Together with the insight into the molecular effects of TNF, the specific inhibition of this cytokine will certainly lead to further novel therapeutic approaches, including the use of small molecule, orally active antagonists. Therefore, the aim of this review is to summarize the recent knowledge on the biology of TNF and the effects of its neutralization in psoriasis.

Immunobiology of TNF-α

  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

The TNF super family and its receptors

TNF-α is a member of a growing family of cytokines referred to as the TNF superfamily, comprising at least 20 peptides (Table 1) (adapted from The Centre for Human Genetics, University College London), which include common mediators such as lymphotoxin (LT), Fas ligand (FasL), and CD40 ligand (CD40L). Other members of the superfamily have such esoteric names as APRIL (a proliferation-inducing ligand), TRAIL (TNF-related apoptosis-inducing ligand), TWEAK (TNF-like and weak inducer of apoptosis), BLyS (B-lymphocyte stimulator), BAFF (B-cell-activating factor belonging to the TNF family), and RANK ligand (receptor activator of NF-κB ligand, RANKL).

Table 1.  Members of the tumor necrosis factor superfamily and their known receptors
TNF superfamilyTNF receptor superfamily
Systematic nameFunctional nameSystematic nameFunctional name
  1. Adapted from The Center for Human Genetics, University College London. TNF, tumor necrosis factor.

TNFSF5CD40L, CD154, TRAP, gp39, IMD3, HIGM1TNFRSF5CD40, p50
TNFSF6FasL, Apo 1LTNFRSF6Fas, Apo 1, CD95, APT1
TNFSF7CD27L, CD70TNFRSF7CD27, Tp55, S152

In recent years, there has been an effort to standardize the nomenclature for this ever-enlarging superfamily of cytokines. For example, a TNF conference in 1998 proposed a nomenclature similar to that accepted for the chemokine family, composed of the term TNFSF for TNF superfamily or TNFRSF for TNF receptor superfamily and a number. TNF-α was assigned TNFSF2, while RANKL was assigned TNFSF11 (Table 1).

Although there are clearly advantages to this nomenclature, there are some distinct disadvantages, the most notable being its complexity, especially to individuals not familiar with the field. Because this standardized nomenclature has not been universally accepted, we have chosen to retain the more classical functional nomenclature in this review, recognizing its inherent limitations.

The functions of many of these proteins are still not known, and their roles in the pathogenesis of inflammatory skin diseases have not been fully established. The activities mediated by this superfamily of ligands and their receptors are extremely diverse and range from the destruction of tissues to the orchestration of lymphoid organogenesis. However, there are some commonalties among members of this superfamily. For example, members of the TNF superfamily, with the exception of LT, are primarily homotrimeric proteins and exist mainly as membrane-associated forms. LT can exist as either a secreted LT-α homotrimer that binds to the two TNF receptors, or as a membrane-associated LT-β heterotrimer composed of LT-α and LT-β chains. This latter LT binds to a unique LT-β receptor. As a general rule, members of the TNF superfamily are also mostly involved in the regulation of cell proliferation and apoptosis, although several of the members, including TNF-α, LT-α, FasL, CD30 ligand (CD30L), and CD40L, also have proinflammatory properties, primarily through the induction of NF-κB. In contrast, TRAIL appears to be unique in that it is a prototype inhibitor protein that antagonizes autoimmune inflammation, in part, by blocking cell-cycle progression.

History of TNF-α

The biological activity of TNF-α was first described by Lloyd Old and his colleagues in the mid 1970s with the identification of a macrophage-derived product that caused hemorrhagic necrosis of solid tumors (6). The amino acid sequence of TNF-α has been known for approximately 20 years, with its cDNA having been cloned and sequenced in the mid 1980s by several laboratories nearly simultaneously (6–8).

Within a few years of its cloning, the principal biological activities of TNF-α had been identified and reproduced using purified or recombinant protein, including (1) the ability to cause hemorrhagic necrosis of tumors, tissue injury, and shock through the proinflammatory properties of TNF-α on the vascular endothelium; (2) the ability to induce apoptosis in some cancerous or transformed cell lines and in lymphocyte and epithelial cell populations; and (3) the ability to alter intermediate substrate and energy metabolism and induce cachexia (Table 2).

Table 2.  Some biological effects of tumor necrosis factor-α[adapted from (167)]
Immune cellsNonimmune cellsIn vivo
  1. TNF, tumor necrosis factor; CNS, central nervous system; GM-CSF, granulocyte–macrophage colony-stimulating factor; IL, interleukin; ICAM, intercellular adhesion molecule; VCAM, vascular cell adhesion molecule.

Monocytes/macrophagesVascular endothelial cellsCNS
 Activate/autoinduce TNF-α Modulate angiogenesis Fever
 Induce cytokines and prostaglandins Increase permeability Anorexia
 Chemotaxis and transmigration Antibrinolytic/procoagulant Altered pituitary hormone secretion
 Stimulate metabolism Suppress proliferation 
 Inhibit differentiation Rearrange cytoskeletonCardiovascular
 Suppress proliferation Induce NO synthasase Shock
  Induce cytokines IL-1, IL-3, G-CSF, and GM-CSF Capillary leak
Polymorphonuclear leukocytes Induce prostacyclin Antithrombosis
 Prime integrin response Induce E-selectin, ICAM, and VCAMGastrointestinal
 Increase phagocytic capacity  Ischemia
 Enhance production of superoxideFibroblasts Colitis
 Increase adherence to extracellular matrix Induce proliferation Hepatic necrosis
  Induce cytokines IL-1, IL-6, and LIF Inhibit albumin expression
Lymphocytes Induce metalloproteases (MMPs) Decrease hepatic catalase
 Induce T-cell colony Suppress respiratory activityMetabolic
 Induce superoxide in B cells Inhibit collagen synthesis Net lipid catabolism
 Induce apoptosis in mature T cells  Net protein catabolism
 Activate cytotoxic T-cell invasivenessAdipocytes Release stress hormone
  Enhance release of free fatty acids Insulin resistance
  Suppress lipoprotein lipase 
 Endocrine system Activate cytotoxicity
  Stimulate ACTH and prolactin Enhance NK-cell function
  Inhibit TSH, FSH, and GH Mediate IL-2 tumor toxicity
  Enhance IL-1 inhibition of steroidogenesis 

Progress in understanding the regulation of TNF-α expression and its signaling pathways has moved with astonishing speed. Beutler reviewed the progress made during this period when the protein was identified and its sequence determined:

‘TNF was born as a purified protein species during an era of technical advancement such as biology had never known. Only a few months elapsed between the time that the protein was first isolated and the time that its entire primary structure was determined by molecular cloning. Only 4 years elapsed between the latter event and the complete determination of its tertiary structure... Never before has information concerning a single protein accumulated with such extraordinary speed’ (9).

Regulation of TNF-α expression

Although it is generally assumed that the primary function of TNF-α is beneficial in activating the innate immune response, inappropriate production of TNF-α leads to inflammation, tissue destruction, and organ injury. Given the primary role that TNF-α plays both in the normal physiologic response to inflammation and in acquired and innate immunity, as well as the substantial pathological consequences associated with its inappropriate synthesis and release, it is not surprising that the expression and activity of TNF-α are tightly regulated at many different levels.

Human TNF-α, which is located on chromosome 6 (Fig. 1), is translated as a 233 amino acid, 26-kDa proprotein that lacks a classic signal peptide (Fig. 2) (10). Newly synthesized proTNF-α is first displayed on the plasma membrane and is then cleaved in the extracellular domain to release the mature monomer through the actions of matrix metalloproteases. The primary enzyme responsible for the processing of cell-associated to secreted TNF-α is TNF-α-converting enzyme (TACE). TACE is an Adamalysin, a member of a class of membrane-associated enzymes that contain both disintegrin and matrix metalloprotease domains (11). This class of enzymes appears critical in the processing of several membrane-associated proteins, including TNF-α, FasL, the TNF receptors, and the epidermal growth factor (EGF) receptor. At the present time, the primary substrates for TACE and other matrix metalloproteases are not completely known. The functions of TACE, however, are not limited solely to the processing of TNF-α, because ablation of the TACE gene is developmentally lethal in the mouse, whereas TNF-α gene ablation results in normal development, growth, and reproduction (12). Rather, TACE appears to be a somewhat promiscuous metalloprotease, active on a number of regulatory proteins including TNF-α, its receptors, and the EGF receptor.


Figure 1. Schematic representation of the tumor necrosis factor-α locus [adapted from (165)].

Download figure to PowerPoint


Figure 2. Processing and intercellular signaling of tumor necrosis factor (TNF)-α. Human TNF-α is synthesized as a 233 amino acid, 26-kDa membrane-associated protein with biological activity. The membrane-associated protein is enzymatically cleaved by TNF-α-converting enzyme (TACE), an adamalysin, to a 157 amino acid, 17-kDa soluble protein that readily homotrimerizes. Binding of homotrimeric TNF-α (either cell-associated or soluble) to its receptors (TNF-RI or TNF-RII) induces oligomerization of the receptors required for signal transduction.

Download figure to PowerPoint

Membrane-associated TNF-α is biologically active and is presumed to mediate the cytotoxic and inflammatory effects of TNF-α through cell-to-cell contact. After proteolytic cleavage by TACE, however, the proprotein is converted to the 157 amino acid, 17.3-kDa secreted protein, which oligomerizes to form the active homotrimer (Figs 2 and 3). The C-terminus of each subunit is embedded in the base of the trimer, and the N-terminus is relatively free of the base structure. Thus, the N-terminus does not participate in trimer interactions and is not crucial for the biological activities of TNF-α. Results from mutational analysis have shown that each TNF-α trimer has three receptor interaction sites located in grooves between the subunits near the base of the trimer structure.


Figure 3. Ligand passing of tumor necrosis factor (TNF)-α. Because of steric hindrance, cell-associated TNF-α may remain tightly associated with TNF-RII, stimulating it to transduce a signal (panel A). With trimeric soluble TNF-α at low concentrations, TNF-RII may serve to pass TNF-α to TNF-RI (panel B) (51,52).

Download figure to PowerPoint

TNF-α is produced by numerous cell types that include immune cells (B cells and T cells, basophils, eosinophils, dendritic cells, NK cells, neutrophils, and mast cells), non-immune cells (astrocytes, fibroblasts, glial cells, granuloma cells, keratinocytes, neurons, osteoblasts, retinal pigment epithelial cells, smooth muscle cells, and spermatogenic cells), and many kinds of tumor cells (13). However, monocytes and tissue macrophages are the primary cell sources for TNF-α synthesis, at least during the inflammatory response. TNF-α gene expression is stimulated by a wide variety of agents. In macrophages, TNF-α gene expression is induced by biologic, chemical, and physical stimuli that include viruses, bacterial and parasitic products, tumor cells, complement, cytokines [interleukin-1β (IL-1β), IL-2, interferon-γ (IFN-γ), granulocyte–macrophage colony-stimulating factor (GM-CSF), macrophage CSF (M-CSF), and TNF-α itself], ischemia, trauma, and irradiation. In other cells, other stimuli are effective: lipopolysaccharide (LPS) in monocytes, engagement of the T-cell receptor in T lymphocytes, cross-linking of surface immunoglobulin (Ig) in B lymphocytes, ultraviolet light in fibroblasts, and phorbol esters and viral infections in many other cell types (13).

Synthesis of TNF-α is tightly controlled at several levels to ensure the silence of the TNF-α gene in the absence of exogenous stimulation, or in tissues that are not programmed to synthesize the protein. Therefore, TNF-α is produced only in barely detectable quantities in quiescent cells, but in activated macrophages, TNF-α is one of the major proteins secreted. The gene for TNF-α is one of the ‘immediate-early’ genes induced by a variety of stimuli. Levels of TNF-α mRNA increase sharply within 15–30 min with no requirement for de novo protein synthesis.

However, TNF-α production is also regulated post-transcriptionally. The 3′-region of TNF-α mRNA includes a series of adenosine-uridine (AU) sequences that render the TNF-α message (mRNA) unstable and determine its translational efficiency. These sequences are common in mRNA for several proinflammatory cytokines, and the presence of these sequences assures that TNF-α mRNA cannot be translated, but is rapidly degraded by cytosolic RNAases (14,15). These AU-rich elements are known to be recognition sequences for several RNA-binding proteins, of which only a few have been characterized to date. Some of these proteins appear to be involved in determining TNF-α mRNA stability, while others are involved in translational silencing (16). Han and Beutler showed, for example, that the presence of these 3′-AU-rich elements was responsible for both the suppression of TNF-α mRNA translation in the unstimulated state as well as the derepression that occurs following stimulation of macrophages with bacterial endotoxin (15). The varying capabilities of different tissues to express TNF-α appear to depend as much upon its transcriptional regulation as upon the ability to derepress these translational signals in the AU-rich 3′-untranslated region. These findings also explain why activated complement, IL-1, and TNF-α itself may induce TNF-α mRNA expression, but not translation and protein production, as these stimuli are not efficient at derepressing the translational blockade that exists in resting macrophages. In contrast, agents like bacterial LPS are potent inducers of TNF-α mRNA transcription as well as derepression of the translational blockade. Kontoyiannis and colleagues demonstrated that transgenic mice that lack the AU-rich region in their TNF-α cDNA spontaneously develop chronic inflammatory diseases similar to rheumatoid arthritis and inflammatory bowel disease (17).

Many of the downstream mediators induced by TNF-α also serve to down-regulate TNF-α expression both transcriptionally and post-transcriptionally. For example, induction of corticosteroids and prostanoids by TNF-α down-regulates gene expression, as does the induction of anti-inflammatory cytokines such as IL-10. IL-10 does this by inhibiting IκB kinase and NF-κB activation (18). Corticosteroids also appear to suppress the translation of the TNF-α mRNA. The net result of this feedback loop is an integrated effort to restrict the duration and magnitude of TNF-α expression once induced. This becomes particularly important in chronic inflammatory diseases in which TNF-α production becomes inappropriately sustained, as may occur in the inflamed cutis. Under these conditions, increased endogenous production of prostaglandins, corticosteroids, and IL-10 is generally ineffective at preventing sustained TNF-α expression. In fact, the use of non-steroidal anti-inflammatory drugs (NSAIDs) in rheumatoid arthritis patients may actually increase TNF-α production.

Biological functions of TNF-α

TNF-α was characterized simultaneously as a factor that produced tumor necrosis in vivo and exhibited antitumor activity by inducing cell apoptosis. It has subsequently been recognized that TNF-α: 1) modulates growth, differentiation, and metabolism in a variety of cell types; 2) can produce cachexia by stimulating lipolysis and inhibiting lipoprotein lipase activity in adipocytes and by stimulating hepatic lipogenesis; 3) can initiate apoptosis in malignant or transformed cells, virally infected cells, T lymphocytes, and epithelial cells; and (4) can produce inflammation (Table 2 and Fig. 4).


Figure 4. Pleiotrophic effects of tumor necrosis factor (TNF)-α in different organ systems. ‘+’ indicates induction; ‘–’ indicates inhibition; CNS, central nervous system; PGE2, prostaglandin E2; and IL, interleukin.

Download figure to PowerPoint

Like IL-1, TNF-α is a potent inducer of the inflammatory response and is a key regulator of innate immunity. Inflammatory responses to TNF-α are mediated both directly, through stimulation of the expression of IL-1, and also via more distal proinflammatory cytokines. Secondary mediators that are known to be induced by systemically administered TNF-α include cytokines [IL-1, IL-1Ra, IL-2, IL-4, IL-6, IL-10, IL-12, IL-18, IFN-γ, transforming growth factor-β (TGF-β), leukemia inhibitory factor (LIF), and macrophage migration inhibitory factor (MIF)], hormones (cortisol, epinephrine, glucagon, insulin, and norepinephrine), and various other molecules [acute phase proteins, leukotrienes, oxygen free radicals, platelet-activating factor (PAF), nitric oxide (NO), and prostaglandins]. Some of the other principal biologic effects of TNF-α are listed in Table 2.

There is growing recognition that TNF-α is not only involved in tissue inflammation and injury but also appears to be a prominent ligand for the activation of apoptosis, or programmed cell death. This latter function occurs during normal growth and development but may also result from pathological conditions in which local and systemic production of TNF-α is increased. Exogenous administration of TNF-α to animals with experimentally implanted tumors produces antineoplastic activity secondary to its ability to induce apoptosis in selected tumor cell populations and to disrupt neovascularization of solid cancers. Apoptosis of lymphoid cell populations associated with activation-induced cell death is also mediated in part by TNF-α (19).

TNF-α also plays an important role in the regulation of the Th1 immune response. This is particularly important in rheumatoid arthritis, in which a positive feedback loop linking the autoimmune and the inflammatory or innate immune responses has been established. TNF-α can induce the synthesis of IL-12 and IL-18, two cytokines that are potent inducers of IFN-γ. Therefore, TNF-α, by itself and through up-regulation of IL-12 and IL-18, amplifies the Th1 response, increasing CD4+ T-cell activation and IFN-γ production. In turn, this leads to increased macrophage production of TNF-α and activation of the inflammatory response.

Studies in animal models

Much of what we know about the contribution of TNF-α to the pathologic changes in inflammation-driven autoimmune diseases including rheumatoid arthritis comes from studies in rodent models. In genetically altered mice overexpressing TNF-α, spontaneous rheumatoid arthritis-like lesions developed in the joints, with progressive inflammation, cellular proliferation, and bone destruction. For example, Butler and colleagues demonstrated that mice expressing a human TNF-α transgene modified to increase its expression (by removing the destabilizing AU-rich elements) had increased TNF-α, IL-1β, and IL-6 expression in synovial cells (20). When these mice were backcrossed into the genetically susceptible DBA/1 strain of mice, an accelerated and severe erosive form of arthritis developed. TNF-α expression was primarily restricted to synoviocytes, and surprisingly, there was little evidence of lymphocyte infiltration. More interestingly, the authors constructed a transgenic mouse that expressed only a cell-associated form of TNF-α. By site-directed mutagenesis, the transmembrane and extracellular regions of the TNF-α cDNA were mutated to make the expressed protein resistant to TACE processing. Only the 26-kDa cell-associated form of TNF-α could be synthesized, and it was not further processed to the 17-kDa form. Transgenic mice overexpressing this cell-associated form of TNF-α spontaneously developed a rheumatoid arthritis-like phenotype (21,22). In this case, backcrossing these animals into knockout strains of mice lacking either a functional TNF-RI or TNF-RII reduced the severity of the disease, suggesting that co-operative signaling by cell-associated TNF-α through both TNF receptors was required (21). However, the body of evidence suggests that TNF-α signaling through the TNF-RI was essential for the onset of rheumatoid arthritis and that signaling through the TNF-RII, presumably by cell-associated TNF-α, contributed further to the disease.

In the genetically susceptible DBA-1 strain of mice, increased TNF-α expression has been documented in the synovial lining of the inflamed joints following immunization with type II collagen and concurrent with the onset of symptoms (23–25). Exogenous administration of TNF-α exacerbates the disease in this model, whereas inhibitors of TNF-α (either antibodies or TNF-receptor constructs) appear to prevent disease onset or reduce its severity. The first convincing evidence that blocking TNF-α prevented the onset of disease came almost simultaneously in 1992 from the studies of Thorbecke (26); Williams, Feldmann, and Maini (27); and Piguet (28).

Because TNF-α has the capability to induce the expression of other proinflammatory cytokines, such as IL-1 and several chemokines, some controversy exists over the role that TNF-α plays in the development of autoimmunity. Although there is no doubt that aberrant TNF-α expression in the synovium contributes to the disease progression, it is unclear whether TNF-α directly mediates this process or acts through the expression of more distal cytokines. In rodent models of rheumatoid arthritis, for example, there is some evidence to suggest that the arthritic changes in response to antigenic stimulation and dependence on TNF-α signaling are actually mediated by IL-1. As discussed earlier, TNF-α can induce the expression of a number of other cytokines, including IL-1. There is actually some evidence from rodent models of arthritis that much of the TNF-α-dependent, erosive changes in the joint are mediated by IL-1. In studies by Probert and colleagues, an erosive form of arthritis develops in response to overexpression of human TNF-α (29). These authors have shown, however, that this development of disease is dependent upon IL-1 signaling through its type I receptor. Passively immunizing mice with a monoclonal antibody directed against the IL-1RI prevented the synovial hyperplasia, bone resorption, pannus formation, and inflammatory cell infiltration that accompany overexpression of TNF-α. Interestingly, blocking IL-1 signaling also attenuated the magnitude of the TNF-α response, suggesting that IL-1 may act by relaying and amplifying pathogenic activities provided by other factors, including TNF-α.

These studies by Probert and colleagues suggest that TNF-α and IL-1 signaling in rheumatoid arthritis may be acting in series: that TNF-α induction of IL-1 may contribute significantly to the pathogenesis of arthritis. There is some evidence from human synovial explants to suggest that blocking an endogenous TNF-α response may attenuate IL-1 and other downstream cytokine responses. Ulfgren and colleagues, for example, demonstrated a reduction in either IL-1α or IL-1β expression in seven of eight rheumatoid arthritis patients treated with infliximab (30). However, other data from rodent studies suggest that IL-1 expression may be partially independent of TNF-α (31). In fact, van den Berg has argued that the biological role(s) for TNF-α and IL-1 in rheumatoid arthritis may be distinct. Not only is the expression of each cytokine under independent regulatory control, but TNF-α and IL-1 may contribute to different aspects of the inflammatory and destructive processes (32). This is an evocative premise as it requires that in addition to the known synergistic or additive responses that TNF-α and IL-1 invoke (such as on prostaglandin and NO production), each cytokine may have individual and distinct functions in the panoply of inflammatory and destructive responses to rheumatoid arthritis. Based on this hypothesis, TNF-α induces inflammation directly and through the induction of IL-1, whereas IL-1 is primarily responsible for the bone resorption and cartilage destruction.

There is some evidence from rodent models that different components of the arthritic response can be attributed separately to TNF-α and IL-1. Using a streptococcal cell wall arthritis model, blockade of TNF-α significantly reduced joint swelling, whereas this effect was not found when IL-1 was blocked. In contrast, neutralization of IL-1, but not TNF-α, resulted in a significant decrease in the inhibition of chondrocyte proteoglycan synthesis (33). Similar findings have been seen in the collagen-induced arthritis model (34). For example, both TNF-α and IL-1 blockade ameliorated the swelling and redness of the afflicted joints when applied shortly after onset of arthritis. In addition, both treatments appeared to reduce the magnitude of the systemic inflammatory response, as determined by serum IL-6 concentrations. However, in this model, blockade of IL-1, but not TNF-α, significantly decreased the cartilage and joint destruction. Radiographic analysis of knee and ankle joints revealed that bone erosions were prevented by anti-IL-1 treatment, whereas anti-TNF-α-treated animals exhibited changes comparable to controls. These authors have argued that anti-TNF-α treatments are most effective when the inflammatory response predominates (35). In addition, very recent studies suggest that although TNF-α is important in the development of inflammatory arthritis in rodent models, it is not essential. Campbell and colleagues have recently reported that severe collagen-induced and antigen-induced arthritis can still develop in mice lacking TNF-α, although the severity is reduced (36).

It needs to be strongly reiterated, however, that the most convincing evidence to date that TNF-α plays a central role in the pathogenesis of autoimmune diseases such as rheumatoid arthritis comes from clinical trials with either monoclonal antibodies against human TNF-α or soluble receptor constructs. Inhibition of TNF-α reduces symptomatology, arrests the erosive disease progression, down-regulates the proinflammatory cytokine cascade, diminishes leukocyte trafficking into the joint, reduces angiogenesis, and reverses the hematologic abnormalities that accompany human inflammation-driven autoimmune disease.

TNF-α administration in humans

TNF-α has been administered to humans, both in low doses in human volunteers and in higher doses as part of a regional chemotherapeutic treatment plan for patients with sarcomas and melanoma. In human volunteers, 1.2 mg/kg of recombinant human TNF-α was administered in an effort to elucidate its systemic properties. Not unexpectedly, even with this low dose, the systemic effects were profound. The individuals developed fevers and complained of constitutional symptoms of pain, headaches, myalgia, and nausea. Hematologically, rapid neutropenia followed by neutrophilia was observed, whereas lymphopenia and monocytopenia were sustained (37). These effects occurred in the absence of any evidence of complement activation. Plasma IL-6 levels increased 40-fold, and the individuals developed a hepatic acute phase protein response. Prostaglandin production (elevated 6-ketoPGF-1α) was markedly increased (38). Metabolically, the patients exhibited increased lipolysis and glucose turnover (39). There were also significant effects on the vascular endothelium. Administration of TNF-α induced activation of the fibrinolytic system within 1 h, as well as activation of the coagulation system thereafter (40).

By the late 1980s, there were at least 20 phase 1 clinical trials of recombinant human TNF-α in patients with advanced cancer. Today, TNF-α is still being used experimentally for the regional treatment of sarcomas and melanomas, usually in conjunction with traditional antineoplastic agents. Doses ranging from 0.04 to 13 mg/kg have been administered subcutaneously, intramuscularly, intravenously, or in at least one case, orally. There was no evidence that patients with advanced malignancies had either a greater or lesser response to TNF-α than did individuals without cancer. The most common systemic effects were fever, rigors, headache, fatigue, nausea, and vomiting. These did not appear to be entirely dose dependent and could not be fully suppressed with acetaminophen or NSAIDs such as ibuprofen. Hypotension, in some cases requiring dopamine in addition to crystalloid, was the primary dose-limiting toxicity and was often seen at doses in excess of 5 mg/kg (41). At doses greater than 3.6 mg/kg administered either intramuscularly or subcutaneously, injection-site pain and erythema was dose limiting. Skin ulceration and necrosis at the injection site were not uncommon at this higher dose (42).

In an effort to avoid the systemic toxicities associated with intravenous, subcutaneous, or intramuscular injections, regional administration of TNF-α has been employed. TNF-α was administered in combination with a conventional antineoplastic agent in the regional perfusion of limbs with cancer. Under these conditions, higher doses of TNF-α (3–4 mg) could be administered, and much higher localized concentrations of TNF-α could be achieved. This occasionally led to systemic release of large quantities of protein, with profound effects on the vascular endothelium. Under these conditions, activation of the endothelium was made evident by increased release of soluble selectins and integrins, and hemodynamic instability was often observed (43,44).

TNF receptor signaling

Biological responses to TNF-α are mediated by ligand binding via two structurally distinct receptors (Fig. 3): type I (TNF-RI; p60 or p55; CD120a) and type II (TNF-RII; p80 or p75; CD120b). Both receptors are transmembrane glycoproteins that have multiple cysteine-rich repeats in the extracellular N-terminal domain. TNF-RI and TNF-RII are present on all cell types except erythrocytes. Although the distribution of TNF-RI is more widespread, TNF-RII is present in greater amounts on endothelial and hematopoietically derived cells. TNF-RI expression is constitutive in most cell types, whereas expression of TNF-RII appears to be inducible. Both TNF receptors are subject to proteolytic cleavage by members of the matrix metalloprotease family and are shed from the surface of cells in response to inflammatory signals such as TNF-α ligand-receptor binding. The shed extracellular domains of both receptors retain their ability to bind TNF-α and therefore may act as natural inhibitors of TNF-α bioactivity (45). During chronic and acute inflammatory conditions, the concentrations of both receptors increase dramatically, although the concentration of TNF-RII is generally more labile than is the concentration of TNF-RI. Both shed receptors are cleared by the kidney and excreted in the urine (46), usually immunologically intact (47–49). This shedding of the cellular receptors, their increased plasma concentrations, and their ability to bind TNF-α has led to the hypothesis that shed TNF receptors may serve either as natural antagonists or as delivery peptides (ligand passers) for circulating TNF-α, depending upon their relative concentrations (50).

The two TNF receptors differ in their binding affinities for TNF-α, as well as in their intracellular signaling pathways. Binding of TNF-α to either receptor appears to be one of high affinity, with a derived Kd between 1.9 and 4.2 ¥ 10–11 M; in fact, the overall affinities are probably quite similar (51). However, pulse-chase experiments demonstrate that the kinetics by which TNF-α binds to and is released from the two receptors differ. TNF-α binds to either TNF-RI or TNF-RII with high affinity and with rapid association kinetics (1.1 ¥ 109 and 1.5 ¥ 109 M−1 min−1, respectively). However, these pulse-chase experiments suggest that binding of TNF-α to TNF-RI is nearly irreversible, due to very slow Koff kinetics (0.021/min, T½ = 33 min) whereas TNF-RII had a rapid Koff kinetics (0.631/min, T½ = 1.1 min).

The different kinetics of binding of TNF-α to either TNF receptor has raised speculation that, in vivo, the two receptors have different functions (Fig. 3b). Under some experimental conditions, TNF-RII may serve as a ligand passer, i.e. a means to deliver or pass TNF-α to TNF-RI for signaling when concentrations of TNF-α are low (52). Supporting the hypothesis that TNF-RII functions as a ligand passer is the observation that, under in vivo conditions, the responses to soluble 17-kDa TNF-α are mediated by TNF-RI, rather than by TNF-RII signaling (50).

TNF-α signaling through either TNF-RI or TNF-RII is triggered by the juxtaposition of the intracellular domains of receptor molecules following ligand binding. Binding of TNF-α to a single receptor does not appear to be sufficient to transduce a signal. Rather, the presence of homotrimeric TNF-α is essential for the juxtaposition of the intracellular domains of the TNF receptor, because dimerization and trimerization of the intracellular domains of the TNF receptors are required for signal transduction. It has been suggested that self-associations in the TNF-RI intracellular motifs contribute to the initiation and amplification of the signal.

The intracellular signaling domains of TNF-RI actually share greater homology with the intracellular signaling domains of Fas than they do with those of TNF-RII, particularly with regard to the highly conserved intracellular domain called the death domain (DD). This sequence of approximately 70 amino acids plays a pivotal role in the ability of TNF-α to trigger apoptosis in the cell (Fig. 5). These intracellular death domains recruit other DD-containing molecules and initiate the intracellular signaling cascade. The recruitment of intracellular signaling molecules to the intracellular domain of TNF-RI occurs via intermediate adaptor or docking proteins, most of which have no enzymatic activity of their own. There are several of these docking proteins, including a protein called receptor-interacting protein (RIP), which requires another DD-containing protein, TNF receptor I-associated death domain protein (TRADD). TRADD can also interact with two other proteins, TRAF-1 and TRAF-2, from another family of signal-transducing proteins called TNF receptor-associated factors (TRAFs). TRAF-2 is an intermediary in the activation of NF-κB by TNF-α. TRADD lies at the bifurcation of the apoptotic and proinflammatory signaling pathways of TNF-α (Fig. 5).


Figure 5. Intracellular signaling pathways of tumor necrosis fac- tor (TNF) receptor type I. The intracellular signaling pathways for TNF-RI are shared with Fas/Apo 1. Binding of homotrimeric TNF-α to TNF-RI leads to recruitment of several intracellular signaling proteins, including TRADD and TRAF-2. These two proteins represent the bifurcation of TNF-RI signaling, with the apoptotic signal delivered through caspase-8 and activation of proinflammatory properties delivered through NIK and NF-κB.

Download figure to PowerPoint

The death domain of TNF-RI is not the sole region in the cytoplasmic domain that is involved in signal transduction. Upstream of the DD in the membrane proximal region of TNF-RI, three proteins bind to TNF-RI and are involved in signal transduction: factor associated with neutral sphingomyelinase activation (FAN), TNF-related activation protein-1 (TRAP-1), and TNF-related activation protein-2 (TRAP-2). FAN appears to play a role in the activation of neutral sphingomyelinase responsible for the generation of ceramide. The functions of TRAP-1 and TRAP-2 are presently not known, although TRAP-2 may be involved in the regulation of proteasomal function.

TNF-RII may also participate in the proinflammatory signal of TNF-α via TRAF-2 (Figs 5 and 6). Some investigators, however, have observed that TNF-RII agonists are able to induce apoptosis (53,54), despite the lack of a TRADD/Fas-associated death domain protein (FADD) binding region as found in TNF-RI. Induction of apoptosis by TNF-RII does not share the same pathways as TNF-RI but seems to rely on the induction of TRAF-2. As with TNF-RI, TNF-RII has been found to associate with the C-terminus of TRAF-2, which mediates activation of NF-κB. A protein kinase, NIK, which binds to TRAF-2 and includes NF-κB activation through the sequential phosphorylation of MAP kinases, has been described (55).


Figure 6. Intracellular signaling pathways of tumor necrosis factor (TNF) receptor type II. Unlike TNF-RI, signaling through TNF-RII does not appear to involve death domains and the activation of caspase-dependent apoptosis. However, ligation of TNF-RII leads to activation of NIK- and NF-κB-dependent signaling.

Download figure to PowerPoint

The binding of TNF-α to its receptor can simultaneously initiate several signaling pathways, including those that promote or inhibit apoptosis. Intracellular mechanisms must exist therefore to define which pathway is activated and/or dominant. The apoptosis-inhibition pathway is NF-κB dependent, as shown by studies demonstrating that TNF-α-induced apoptosis of malignant cells was inhibited by simultaneous activation of NF-κB-dependent pathways, and that inhibition of NF-κB markedly increased the apoptotic response to TNF-α in several malignant tumor cell lines (56,57). The selection of the dominant pathway appears to rest with a ‘molecular switch,’ which acts in part through the intracellular concentration of cell-signaling intermediates. This has been best shown in T cells, in which the intracellular concentration of RIP determines whether TNF-RII signaling occurs through apoptotic or NF-κB-dependent pathways (58). Increases in the intracellular concentration of RIP, induced by IL-2, triggered cell death. Under the same conditions, depletion of RIP reduced the susceptibility of the cell to TNF-α-dependent apoptosis. These findings suggest that the signaling outputs are regulated by intracellular factors. They may also help explain some of the conflicting data regarding the proapoptotic vs. inflammation-activating effects of TNF-α.

It should be noted, however, that much of the data from in vivo and in vitro studies do not account for the in vivo contribution of the individual receptor types in TNF-mediated signaling, particularly in response to secreted 17-kDa TNF-α. For example, in vitro both TNF-RI and TNF-RII can transduce a signal for NF-κB activation in several cell types, whereas in vivo studies have suggested that TNF-RI signaling is primarily responsible for the proinflammatory properties of TNF-α. In fact, in vivo studies by Peschon and colleagues and Nowak and colleagues using TNF-RII knockout mice suggest that TNF-RII may function at times like a decoy receptor, because mice lacking a functional TNF-RII will often manifest an exaggerated inflammatory response (59,60). In addition, studies in baboons have shown that TNF-α muteins with specificity for TNF-RI are proinflammatory, whereas TNF-RII agonists are not (61,62). These primate data are consistent with earlier studies performed using transgenic mice or receptor-specific antagonists. For example, antibodies that prevent TNF-α binding to TNF-RI, but not TNF-RII, protected mice from lethal endotoxic shock but blocked development of a protective response against Listeria monocytogenes infection (63). Similarly, transgenic mice lacking a functional TNF-RI are more resistant to TNF-α but are more susceptible to infection by L. monocytogenes (64,65). Although TNF-RII-deficient mice exhibit normal T-cell development and activity, these animals are also more resistant to TNF-α-induced death, suggesting that TNF-RII may have no intrinsic proinflammatory properties of its own but can potentiate the actions of TNF-RI (59,66). Based on observations that in vivo inflammatory responses to soluble 17-kDa TNF-α are mediated primarily by TNF-RI and that TNF-α binding to TNF-RII is associated with very rapid on-off kinetics, it was postulated that TNF-α signaling was mediated primarily by binding of the 17-kDa ligand to TNF-RI. The concept that TNF-α signaling of inflammation and apoptosis in vivo occurs principally through TNF-RI probably requires some modification (67). Grell and associates have demonstrated that the secreted and cell-associated forms of TNF-α have markedly different affinities for the two TNF receptors (67). They propose that the principal ligand for TNF-RI is the 17-kDa secreted form of TNF-α, whereas cell-associated TNF-α is the primary signaling ligand for TNF-RII. The on-off kinetics of 17-kDa TNF-α with the type II receptor are very fast and thus, at low TNF-α concentrations, TNF-RII may serve as a ligand passer and increase TNF-α binding to TNF-RI. Conversely, because of the close juxtaposition of 26-kDa cell-associated TNF-α to TNF-RII that occurs during direct cell-to-cell contact, TNF-α/TNF-RII complexes may be generated with increased stability and signaling potential (Fig. 3b). Steric hindrance by cell-associated TNF-α would prevent ligand passing from TNF-RII and permit signal transduction to occur. Based on this hypothesis, these investigators propose that cell-associated TNF-α is the prime physiologic activator of TNF-RII, implying that TNF-RII contributes to the juxtacrine TNF-α response in tissues, as occurs in experimental hepatitis and rheumatoid arthritis (21,68). Along these same lines, the investigators have demonstrated that overexpression of human TNF-RII can induce an exaggerated inflammatory response in several organs, suggesting that signaling through this receptor has the potential to directly induce tissue damage (69).

The 17-kDa secreted TNF-α (and not the 26-kDa cell-associated form) is primarily responsible for mortality in endotoxin- or bacteremia-induced shock (59,60,62), and this occurs primarily through TNF-RI signaling. Conversely, synovial inflammation and joint erosion appear to be dependent, at least in part, on cell-associated TNF-α signaling, with involvement of TNF-RII. Using a novel transgenic mouse that expresses only a membrane-associated form of TNF-α, Kollias and co-workers demonstrated that expression of the 26-kDa form of TNF-α was adequate by itself to induce arthritis (21).These animals spontaneously developed a pattern of arthritic lesions similar to human rheumatoid arthritis at about 6–8 weeks of age (22). In addition, Williams and colleagues noted that treatment of rheumatoid synovial explants with a matrix metalloprotease inhibitor blocked the processing of TNF-α and stabilized TNF-α receptors on cell membranes but did not affect IL-1, IL-6, or chemokine release (70). In contrast, an antibody against TNF-α blocked the downstream induction of these other proinflammatory cytokines. These results with an antibody against TNF-α are very similar to the results seen with an antibody against the IL-1 receptor, which prevented all aspects of the disease. The data suggest that cell-associated forms of TNF-α, in synovial explants not inhibitable with matrix metalloprotease inhibitors, may contribute to the local production of other proinflammatory cytokines, like IL-1, and still contribute to disease progression.

These findings emphasize the complexity of the TNF-α-signaling system and the multiple levels at which TNF-α signaling is regulated. Not only is the expression of TNF-α tightly controlled at the level of gene transcription and subsequent translation, but also its processing from a cell-associated to a secreted form is regulated by protease activity. The secreted and cell-associated forms of TNF-α may serve independent functions through their differing capacities to bind and transduce a signal through the two receptors. TNF-α signaling is also antagonized or aided by circulating extracellular domains of the TNF receptors, which, depending upon their concentration and the concentration of TNF-α, may serve as either inhibitors or ligand passers. Finally, the distribution of receptors on the target cells ultimately determines the responsiveness of a given tissue to TNF-α.

Studies on the role of TNF-α in inflammation-driven autoimmune diseases

Both animal and human studies have strongly implicated TNF-α in the pathogenesis of rheumatoid arthritis, a prototypical inflammation-driven autoimmune disease. Early documentation identified increased TNF-α expression in the inflamed joints of animals with experimentally induced arthritis and in patients with rheumatoid arthritis. Much of this literature has recently been reviewed in depth by Feldmann (71). To date, anti-TNF therapies have been tested clinically in at least 18 different inflammation-related diseases (Table 3).

Table 3.  Anti-TNF-α-based therapies in various diseasea
  • TNF, tumor necrosis factor; HIV, human immunodeficiency virus.

  • a

    Some studies are reports of non-controlled experience in small number of selected patients.

Rheumatoid arthritisEtanercept; pegsunercept; infliximab; adalimumab; CDP870
Psoriatic arthritisEtanercept; infliximab; oncercept
PsoriasisEtanercept; infliximab; oncercept; adalimumab
Adult Still's diseaseInfliximab
Severe active ulcerative colitisInfliximab
SpondyloarthropathiesEntanercept; infliximab
Ankylosing spondylitisEntanercept; infliximab
Behçet's syndromeInfliximab
Crohn's diseaseInfliximab; onercept; adalimumab; CDP870
Orofacial Crohn's diseaseInfliximab
UveitisEntanercept; infliximab
HIV-1-associated psoriatic arthritisEtanercept
Graft-vs.-host diseaseEtanercept
Advanced heart failureEtanercept
Common variable immunodeficiencyEtanercept
Wegener's granulomatosisEtanercept

The presence of different cytokines in the synovial fluid of patients with rheumatoid arthritis was examined as early as 1988. TNF-α (72) as well as IL-1, TGF-β, and platelet-derived growth factor (PDGF) were detected in large quantities in synovial fluid from patients with rheumatoid arthritis, whereas T-cell-produced cytokines (IL-2, IL-4, and IFN-γ) were generally absent or present in small quantities [reviewed by Feldmann and colleagues (73)]. Immunohistochemical localization of TNF-α demonstrated predominant expression in macrophages (74,75). Similar studies using ex vivo implants from patients with rheumatoid arthritis demonstrated increased TNF-α expression, and biologically active TNF-α has been recovered from synovial membrane cultures.

Studies by Alvaro-Gracia and Firestein directly examined the pattern of cytokine expression in synovial tissues from patients with rheumatoid arthritis. They observed a high frequency of cells (>5%) expressing IL-1β and TNF-α, whereas IFN-γ expression was minimal (74). Colocalization studies confirmed that macrophage-lineage cells were the primary source of the increased TNF-α and IL-1 expression, whereas IL-6 was expressed primarily by cells of non-T lineage. The same investigators observed that both TNF-α and IL-1 stimulated synoviocyte proliferation, collagenase activity, and GM-CSF activity (76,77). Not only was IFN-γ not detected in synovial cells from patients with rheumatoid arthritis, but added IFN-γ actually interfered with TNF-α- and IL-1-mediated proinflammatory responses in the synovium (78). Interestingly, Ulfgren and colleagues demonstrated that TNF-α, IL-1α, and IL-1β production in synovial biopsy explants from patients with rheumatoid arthritis was reduced by in vivo administration of an anti-TNF-α antibody (30). These studies suggest that blocking TNF-α may not only down-regulate the expression of IL-1 but also contribute to the sustained expression of TNF-α.

Similarly, TNF receptor expression is up-regulated in active rheumatoid arthritis tissues, especially in the synovium in areas adjacent to tissue erosion. Increased levels of shed TNF receptors have been detected in the serum of patients with rheumatoid arthritis, and more importantly, the increases in concentrations of the shed receptors appear to correlate with disease activity (79–81).

More relevant evidence suggesting that TNF-α directly contributes to the pathogenesis of the disease comes from studies in animals with experimentally induced disease. The first studies showing that TNF-α had the potential to induce proteoglycan resorption and block proteoglycan synthesis came from the studies of Saklatvala in 1986 (82). Since then, there has been a considerable body of research demonstrating that intra-articular administration of TNF-α can recapitulate many of the pathological changes seen with rheumatoid arthritis.

Expression of TNF in psoriasis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

Overexpression of TNF-α has been demonstrated in psoriasis for well over a decade (83–89). In lesional psoriasis skin, and to a lesser extent in uninvolved psoriasis skin, TNF-α was found to be distributed throughout the epidermis and was also specifically localized to the upper dermal blood vessels (83). Other authors have localized TNF-α in psoriatic lesions with dermal macrophages in the papillary dermis and focally by keratinocytes and intraepidermal Langerhans' cells but did not find it to be significantly expressed in endothelial cells, mast cells, or dermal Langerhans' cells (88). Importantly, the two receptors for TNF-α, TNF-RI, and TNF-RII are differentially expressed in the skin: in normal skin, and uninvolved and lesional skin from psoriasis patients, the p55 TNF-RI is associated with epidermal keratinocytes and a network of upper dermal dendritic cells. In lesional psoriasis skin, TNF-RI was detected in the parakeratotic stratum corneum and increased expression of TNF-RI was found in association with upper dermal blood vessels. In contrast, staining of the p75 TNF-RII in normal skin was found to be restricted to eccrine sweat ducts and dermal dendritic cells, and was absent from the epidermis (83).

However, overexpression of this proinflammatory cytokine had been sufficiently demonstrated in skin lesions but not in peripheral blood of psoriatic patients. Therefore, we determined the TNF-α mRNA expression in peripheral blood mononuclear cells (PBMCs) from psoriatic patients and healthy controls. Using a semiquantitative competitive reverse transcriptase-polymerase chain reaction, a significantly higher TNF-α mRNA expression was found in the PBMCs from psoriatic patients when compared to healthy controls. We also wondered whether the elevated TNF-α gene expression resulted in enhanced plasma levels of bioactive TNF-α protein. Using a highly sensitive enzyme-linked immunosorbent assay (ELISA) recognizing only the bioactive TNF-α trimeric molecule, we found slightly but significantly elevated TNF-α plasma concentrations in psoriatic patients (90).

High levels of TNF-α have also been detected in psoriatic arthritis. The pattern and the expression levels of proinflammatory Th1 cytokines and the monokines TNF-α and IL-1β were found to be similar in synovial tissue of psoriatic arthritis when compared to rheumatoid arthritis (91,92).

Strategies for the inhibition of TNF

  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

As noted above, two distinct TNF receptors have been identified: the 55-kDa or p55 receptor type I (TNF-RI) and the 75-kDa or p75 receptor type II (TNF-RII). Both TNF-RI and TNF-RII exist as cell-surface and soluble forms, and both forms bind TNF, although with different affinities (50). TNF cell-surface receptors are present on nearly all cell types, including macrophages, lymphocytes, and neutrophils. TNF-α must bind to two or three cell-surface receptor molecules for signal transduction to occur. Numerous biological effects of TNF-α are mediated by the intracellular signaling of the high-affinity TNF-RI receptor (51) as well as the low-affinity TNF receptor (59). Signal transduction occurs when TNF-α binds to and dimerizes two or three receptors of either the TNF-RI or TNF-RII subtype on the cell surface (93). Naturally occurring TNF-α inhibitors, consisting of the full-length 4-domain or truncated forms of the extracellular region of TNF-RI, are referred to as TNF-binding proteins (TNFbp) or soluble TNF receptor (sTNF-R) (94). These molecules have been found in the tissue, serum, synovial fluid, and synovial explant cultures obtained from patients with active rheumatoid arthritis.

Because of the natural properties of these receptors, biopharmaceutical investigators over the last 10 years have designed a wide array of biological agents to inhibit TNF-α. These include (Fig. 7) 1) a TNF type II soluble receptor fusion protein (etanercept, Enbrel®, Amgen/Wyeth, Thousand Oaks, CA, USA) (95,96); 2) a specific antihuman TNF-α chimeric (mouse × human) monoclonal antibody (infliximab, Remicade®, Johnson & Johnson/Schering-Plough, New Brunswick, NJ, USA) (97–99); 3) a humanized antihuman TNF-α monoclonal antibody (adalimumab, D2E7, humira, Abbott Laboratories/Cambridge Antibody, Abbott Park, IL, USA) (100,101); 4) a human TNF type I (p55) soluble receptor (r-hTBP-1, onercept, Serono, Geneva, Switzerland); 5) a pegylated form of soluble TNF type I (p55) receptor (PEG-sTNFR1, pegsunercept, Amgen, Thousand Oaks, CA, USA), and 6) a pegylated humanized anti-TNF antibody fragment (CDP870, Celltech/Pfizer, Slough, UK). Another TNF type I soluble receptor (p55) fusion protein, lenercept (Roche, Basel, Switzerland), has demonstrated short-term efficacy in rheumatoid arthritis in European and North American phase 2 clinical trials but was highly immunogenic upon long-term dosing (102–106). The antibodies generated by lenercept dosing were bound to Fc receptors but were not detectable to sTNF-RI and had neither neutralizing nor antagonistic properties (102). Furthermore, a series of problems in its development let to its eventual abandonment.


Figure 7. Structure of tumor necrosis factor inhibitors currently being tested in the clinic [adapted from (71)].

Download figure to PowerPoint

TNF-binding protein (TNFbp) (Amgen), a dimeric, pegylated form of the high-affinity p55 soluble TNF type I receptor produced in Escherichia coli, has also been studied preclinically (107,108) and in clinical trials in rheumatoid arthritis (94). The immunogenicity of TNFbp negatively impacted the clearance rate of the molecule and reduced the serum half-life in the phase I/II clinical trial; it was therefore determined to be unsuitable for a chronic indication (94).

Etanercept is Food and Drug Administration (FDA)-approved for rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis (AS). The application for FDA approval for the use in psoriasis has been filed in July of 2003. Infliximab has FDA approval for rheumatoid arthritis and Crohn's disease (CD) and now in phase III clinical trials for psoriasis. Adalimumab has been approved by the FDA for use in rheumatoid arthritis and is currently being tested in phase II trials for psoriasis. Pegylated soluble TNF type I soluble receptor (pegsunercept) (45) is currently in phase III clinical trial for rheumatoid arthritis, and r-hTBP-1 onercept (109,110) is being tested clinically in phase III trials for psoriasis and psoriatic arthritis. CDP870 is now in phase III trial for rheumatoid arthritis and CD (71). (Tables 3 and 4)

Table 4.  Development and approval status of injectable anti-TNF-α therapies in clinical use
Drug (trade name)Company/partnerBiological form/ administrationStatus with US FDAApproval for indication/ development stageDevelopment stage in psoriasis
  1. Fab, fragment antibody binding; Fc, fragment constant; IgG, immunoglobulin G; TNFR1, TNF receptor type I; TNFR2, TNF receptor type II.

Etanercept (Enbrel)Amgen/WyethSoluble TNFR2 coupled to Fc portion of IgG/subcutaneousApproved November 1998Rheumatoid arthritis Psoriatic arthritisUS FDA filing July 2003
    Ankylosing spondylitis 
Infliximab (Remicade)Johnson & Johnson/ Schering-Plough/ Tanabe SieyakuMouse-human chimeric anti-TNF monoclonal antibody/intravenousApproved November 1999 August 1998Rheumatoid arthritis Crohn's diseasePhase III
Adalimumab, D2E7Abbott Laboratories/Human anti-TNFApproved DecemberRheumatoid arthritisPhase II
(Humira)Cambridge Antibodymonoclonal antibody/subcutaneous2002  
r-hTBP-1SeronoHuman soluble TNFR1/ClinicalPhase III for psoriasisPhase III
(Onercept) subcutaneous   
PEG-s TNFR1AmgenPegylated form of soluble TNFR1/ClinicalPhase III forNA
(Pegsunercept) subcutaneous rheumatoid arthritis 
CDP870Celltech/PfizerPegylated Fab of humanizedClinicalPhase III forNA
  antibody rheumatoid arthritis and 
  CDP571/subcutaneous Crohn's disease 

Clinical effects of anti-TNF therapy in psoriasis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

The monoclonal mouse human chimeric anti-TNF-α antibody infliximab (Remicade) was initially used in a psoriatic patient who was also suffering from inflammatory bowel disease. A dramatic improvement in the skin lesions was observed as soon as 2 weeks after a single infusion (111). Good activity has also been reported in combination with methotrexate (112). More recently, efficacy and safety of infliximab monotherapy was demonstrated in a controlled randomized trial. Thirty-three patients with moderate-to-severe plaque psoriasis were randomized into three groups and received either placebo or the antibody at two dosages (5 mg/kg or 10 mg/kg) intravenously at weeks 0, 2, and 6. Eighty two percent in the infliximab 5 mg/kg group and 73% in the 10 mg/kg group had at least 75% improvement in the psoriasis activity severity index (PASI) (18% patients in the placebo group). The PASI score is used to define psoriasis severity and takes into account the localization, extent, and severity of the skin signs such as erythema, scaling, and induration.

The mean PASI decreased from 22.1 to 3.8 (5 mg/kg) and 26.6–5.9 (10 mg/kg) at week 10. The mean time to response was 4 weeks (113) (Table 5). Immunohistochemical analysis of the psoriatic lesions in the patients enrolled in this study demonstrated a rapid and dramatic decrease in epidermal inflammation and normalization of keratinocyte differentiation that preceded maximal clinical response. Importantly, infliximab concentrations could be detected in the majority of patients through week 14 (114).

Table 5.  Clinical studies of infliximab in psoriasis and psoriatic arthritis
  1. i.v., intravenous.

Moderate to severe plaque psoriasisRandomized, double-blind, placebo-controlledInfliximab, 5 mg/kg (n = 11) or 10 mg/kg (n = 11) i.v. weeks 0, 2, and 6 Placebo (n = 11)33Infliximab monotherapy was well tolerated and effective for the treatment of psoriasis;(113)
    75% reduction in PASI in 82%(114)
    of the 5 mg/kg and 73% of the 10 mg/kg groups 
    by or before week 10 compared to 18% in the placebo group 
Moderate to severe psoriasis vulgaris‘SPIRIT’, multicenter, randomized, double-blind, placebo-controlledInfliximab, 3 mg/kg (n = 99) or 5 mg/kg (n = 99) i.v. weeks 0, 2, and 6 Placebo (n = 51)24975% reduction in PASI in 88% of the 5 mg/kg and 72% of the 3 mg/kg groups at week 10 compared to 6% in the placebo group(115) (116)
Psoriasis vulgarisOpen-labelInfliximab fixed dose of 200 mg5Mean decrease of 71% in PASI by 6(117)
  i.v. per treatment (2-3 mg/kg at weeks 0, 2, and 6) weeks [baseline PASI of 20 (n = 4) dropped to 6 (n = 4)] 
Severe pustularOpen-labelInfliximab 5 mg/kg1Complete pustule resolution within 48 h(118)
psoriasis of von Zumbusch single infusion   
Recalcitrant psoriasisOpen-labelInfliximab 5-10 mg/kg2Rapid and complete clearing of(119)
  single infusion recalcitrant plaques and erythroderma and 
    improved symptoms of psoriatic arthritis 
Active psoriasis andOpen-labelInfliximab, 3 mg/kg9Multiple infusions of infliximab were well(120)
psoriatic arthritis (PsA) (n = 9) i.v. 5 times over 22 weeks tolerated and effective in patients with 
    active psoriasis and PsA; ACR20/50/70 
    achieved in 89%/56%/22%; PASI improved from 19 to 5 
Psoriatic arthritis unresponsive to disease modifying antirheumatic drugs (DMARDs)Open-label, compassionate-useInfliximab, 5 mg/kg i.v. at weeks 0, 2, and 6 (n = 10); individualized dosing after week 10; patients continued current therapy until week 1010Infliximab was well tolerated, effective and safe in PsA; ACR20 in all patients by week 2; ACR70 at week 10(121)
Psoriatic arthritisOpen-labelInfliximab, 5 mg/kg i.v. at6Rapid improvement of psoriatic(122)
unresponsive to weeks 0, 2, and 6 skin lesions and arthritis in all patients 
Active psoriatic arthritis‘IMPACT study‘ Randomized, double-blind, placebo-controlled, multicenterInfliximab, 5 mg/kg i.v. with or without methotrexate at weeks 0, 2, 6, and 14, followed by open-label treatment with 5 mg/kg every 8 weeks with follow up through week 50101Average reduction in PASI among the 39 patients who qualified for PASI was 81% in the treatment group compared to increase of 36% in the placebo group; 67% of patients treated with infliximab(123)
    achieved improvement of at least 75% compared to 
    0% in placebo group; ACR20 was met in 69% of 
    infliximab-treated patients compared to 8% in the placebo group 

Similar results of infliximab in psoriasis were achieved in the SPIRIT study that enrolled 249 patients with moderate-to-severe psoriasis. At week 10, 88 and 72% of patients treated with 5 and 3 mg/kg of infliximab, respectively, experienced a >75% improvement from baseline in PASI compared to 6% of these treated with placebo. Among those treated with 5 mg/kg infliximab, nearly half achieved a >75% improvement in PASI as early as week 4 and 58% achieved 90% improvement in PASI by week 10 (115,116).

In a small open-label study with a fixed low dose of infliximab (200 mg intravenous) at weeks 0, 2, and 6 psoriasis patients with a base line PASI of 20 achieved a mean decrease of 71% to a PASI of six by 6 weeks (117).

Infliximab has also demonstrated clinical benefit in severe pustular psoriasis of von Zumbusch type (118) and in recalcitrant psoriasis (119).

In a study that enrolled nine patients with psoriatic arthritis and psoriasis, infliximab was given intravenously over 22 weeks at a dose of 3 mg/kg. The American College of Rheumatology (ACR) criteria for 20, 50, and 70% improvement (ACR20/50/70) was achieved in 89, 56, and 22%, respectively, and mean PASI score improved from approximately 19 to 5 (120).

Treatment with infliximab was also shown to be effective in patients with psoriatic arthritis that are unresponsive to disease modifying antirheumatic drugs (DMARDs). Out of 10 patients enrolled in an open-label study with 5 mg/kg of intravenous infliximab over 10 weeks, all patients achieved a 20% improvement according to the ACR criteria (ACR20) by week 2. Eight patients had improved 70% (ACR70) at week 10 (121). Improvement of psoriatic skin lesions was also observed in six patients with progressive joint disease and psoriatic skin lesions unresponsive to methotrexate therapy (122). In a larger randomized, double-blind, placebo-controlled study enrolling 101 patients with active psoriatic arthritis, patients received either infliximab 5 mg/kg or placebo administered with or without methotrexate at weeks 0, 2, 6, and 14, followed by an open-label treatment with 5 mg/kg every 8 weeks with follow-up through week 50. The average reduction of PASI was 81% in the infliximab group compared to an average increase of 36% in the placebo group. Moreover, 67% of the patients treated with infliximab achieved an improvement of at least 75%, compared to 0% of the patients treated with placebo. These results were maintained through week 50 (116,123).

Similar promising data were reported on the soluble TNF-α receptor etanercept (Enbrel), in the treatment of psoriatic arthritis and psoriasis. Sixty patients with psoriatic arthritis and psoriasis were included in a placebo-controlled study. The treatment (25 mg by subcutaneous injection twice weekly for 12 weeks) was well tolerated with no serious side effects reported. A distinct improvement of the joint involvement (ACR criteria improvement; ACR20) was observed in 73% of the patients treated, compared to 13% of the patients in the placebo group. In addition, the PASI fell by over 75% in 26% of the actively treated patients with psoriasis after 12 weeks vs. in 0% of the placebo group. Overall, a median PASI improvement of 46% in etanercept-treated patients was achieved vs. 9% in the placebo-treated patients (124) (Table 6).

Table 6.  Clinical studies of etanercept in psoriasis and psoriatic arthritis
Psoriasis,Randomized, double-blind,Etanercept, 25 mg subcutaneous60Etanercept was safe, well tolerated(124)
psoriaticplacebo-controlled2 times/week × 12 week (n = 30) and effective in improving signs and symptoms 
arthropathy Placebo (n = 30) of psoriatic arthritis; ACR20 in 73 vs. 13% in 
    placebo group at 12 weeks; median 
    improvement in PASI was 46 vs. 9% 
    in placebo group; 75% decrease in PASI 
    achieved in 26% of patients 
    after 12 weeks vs. 0% in placebo group 
PsoriasisOpen-labelEtanercept × 36 week (n = 19)38Etanercept was well tolerated and(125)
 extensionPlacebo × 12 week -> effective in reducing 
 of above studyEtanercept × 24 week (n = 19) symptoms of psoriasis 
Psoriasis, psoriaticRandomized, double-blindEtanercept, 25 mg subcutaneous205Etanercept was well tolerated(168)
arthropathy 2 times/week × 24 week (n = 101) and effective in improving arthritis(169)
  Placebo (n = 104) and psoriasis; ACR20 in 59% vs. 15% in 
    placebo group at 12 weeks; 
    median improvement in target 
    psoriatic lesions 33% vs. 0% in placebo group 
Active but clinicallyRandomized, double-blind,Etanercept, 50 mg subcutaneous652Etanercept treatment was highly efficacious(129,130)
stable psoriasisplacebo controlled, multicenter2 times/week × 24 week (n = 164); 25 mg subcutaneous in psoriasis with sustained and dose-dependent efficacy; efficacy; 
  2 times/week × 24 week (n = 162); 75% decrease in PASI in 49% of the 
  25 mg subcutaneous 1 time/ week × 24 50 mg ×2/ week group 
  week (n = 160); at week 12 (59% at week 24) vs. 4% 
  Placebo (n = 166; placebo until in the placebo group; 
  week 12, then 25 mg subcutaneous 2 times/week until week 24) adverse events and infections comparable to placebo at week 12 
Moderate to severeRandomized, double-blind,Etanercept, 25 mg subcutaneous112Etanercept monotherapy was(170)
psoriasis, psoriaticplacebo controlled,×2/ week × 24 week (n = 57) Placebo (n = 55) well tolerated; 75% decrease 
arthropathymulticenter  in PASI in 30% vs. 2% in 
    the placebo group at 12 weeks 
    and in 56% vs. 5% in the 
    placebo group at 24 weeks 
Generalized pustularOpen-labelEtanercept, 25 mg, subcutaneous ×2/week × 2 week1Etanercept treatment led to rapid improvement of psoriatic arthritis and(132)
psoriasis   regression of pustular eruption 

In a 24-week, open-label extension of this study, patients who originally received placebo rapidly achieved responses to etanercept that were comparable to responses in the group originally randomized to the active drug; in the group originally on etanercept, efficacy was maintained, and a large proportion of patients decreased or discontinued concomitant prednisolone or methotrexate (125). At the end of the extension study, 82% of psoriatic arthritis patients had achieved the psoriasis arthritis response criteria (PsARC) and 74% the ACR20 criteria (126).

Results from this phase II study were confirmed in a phase III study that enrolled 205 patients with psoriasis and psoriatic arthritis. Patients received 25 mg etanercept (n = 101) or placebo (n = 104) subcutaneously twice weekly for 24 weeks. Concomitant therapy with methotrexate, oral corticosteroids, and NSAIDs was permitted. ACR20 was assessed at 12 weeks and psoriasis involvement was measured by target lesion score, and in a subset of patients with psoriasis involvement >3% by PASI. Etanercept was well tolerated in this patient population. The ACR20 was achieved by 59% of the etanercept group and 15% of the placebo group at 12 weeks. The improvement in arthritis could be maintained with continuing etanercept treatment through the end of the study at 24 weeks. Also, psoriasis was significantly improved with etanercept in this study, with a median improvement in target lesions of 33% in etanercept patients and 0% in placebo patients at 24 weeks. The PASI results were consistent with these results. Recent findings from a open-label extension of this study indicate that etanercept significantly inhibited joint destruction, bone erosion, and joint space narrowing in psoriatic arthritis with the inhibition of joint destruction observed as early as 6 months (127,128).

An even bigger phase III double-blind, placebo-controlled study randomized and dosed 652 patients with active but clinically stable plaque psoriasis involving >10% of the body-surface area (129). Patients received either 50 mg etanercept twice weekly (n = 164), 25 mg twice weekly (n = 162), and 25 mg once weekly (n = 160), or placebo (n = 166). Patients in the placebo group received 25 mg twice weekly from week 12 until the end of the study at week 24. At week 12, a 75% decrease in PASI was reached in 49% of the 50 mg twice weekly group (59% at week 24), in 34% of the 25 mg twice weekly group (44% at week 24), in 14% of the 25 mg once weekly (25% at week 24), and in 4% of the placebo group (33% at week 24; these patients had received 25 mg twice weekly from week 12–24). This study concluded that etanercept treatment was highly efficacious in psoriasis with continued improvement seen through the 24 weeks and a consistent, sustained, and dose-dependent efficacy. The adverse events and infections were comparable to placebo at week 12, and the tolerability and adverse event profile remained similar through 24 weeks of dosing (129).

The efficacy and safety of etanercept (25 mg subcutaneously twice weekly) as a treatment for chronic moderate-to-severe plaque psoriasis were demonstrated in a multicenter, randomized, double-blind, placebo-controlled, 24-week study involving 112 patients. Of the patients receiving etanercept, 30% experienced an improvement in PASI scores of 75% at week 12 vs. 2% in the placebo group. In contrast to the results of the psoriatic arthritis trial, efficacy continued to improve with longer treatment, with 56% of etanercept-treated patients and 5% of those with placebo achieving ≥75% PASI improvement at 24 weeks (130). At the end of this 24-week study, 17 etanercept-treated patients entered an extended observation period off etanercept to evaluate response durability. Relapse or treatment failure was defined as PASI returned to 75% of baseline value, or if the patient initiated systemic psoriasis therapy or if the patient discontinued for any reason. Of the 17 patients off etanercept that were followed up, three patients completed the study through week 52, and 14 patients withdrew because of relapse to 75% of baseline PASI (n = 3), decline to continue (n = 3), start of systemic therapy (n = 7), or disease progression (n = 1). The median time to relapse to 75% of baseline PASI of all patients was 24.3 weeks. The disease relapse was gradual and did not exceed the baseline value and in no patient did plaque psoriasis convert to guttate, erythrodermic, pustular, or other morphologic forms of psoriasis (131).

Furthermore, etanercept has been reported to also be effective in the generalized pustular exacerbation of psoriasis after withdrawal of cyclosporin-A (132). This single report showed that etanercept might also be safe and effective in cases of pustular rebound after withdrawal of immunosuppressive agents.

Also r-hTBP-1 (onercept), the recombinant, unmodified, fully human soluble type I TNF receptor (p55), which acts as an anti-TNF agent has shown good efficacy in the treatment of psoriasis and psoriatic arthritis (109,110). In a 12-week, multicenter double-blind placebo-controlled phase II study for psoriasis, 130 patients with moderate-to-severe psoriasis were treated with either placebo (n = 43), 150 mg three times a week (n = 43), or 100 mg seven times a week (n = 44). Patients treated with onercept at a dose of 150 mg, subcutaneously, three times a week for a period of 12 weeks, showed a significant improvement in their PASI score. After 12 weeks of therapy, 54% of patients receiving onercept 150 mg demonstrated 75% or greater PASI score improvement vs. 12% of patients on placebo. In addition, 74% achieved 50% or greater PASI score improvement (PASI 50) vs. 26% in the placebo group. In the study, patients treated with onercept showed significant improvement in PASI after 2 weeks of treatment compared with placebo. Side effects for onercept-treated patients were similar to those observed in the placebo group. Injection-site reactions were more frequent in the onercept group.

An earlier multicenter double-blind placebo-controlled study of onercept in psoriatic arthritis included 126 patients divided between placebo (n = 42), onercept at 50 mg three times a week (n = 42), and onercept at 100 mg three times a week (n = 42). At baseline, patients were required to have a PASI of eight or more, a body-surface involvement of 5% or more, and at least three active joints. Onercept 50 mg and 100 mg were given subcutaneously three times a week for a period of 12 weeks. The improvement in the PsARC was more favorable at the 100-mg dose. After 12 weeks of treatment at 100 mg, 86% of patients on onercept met the PsARC primary endpoint compared with 45% on placebo. ACR20 was achieved by 67% of onercept patients receiving 100 mg compared to 31% on placebo. Side effects for onercept-treated patients were similar to those observed in the placebo group. Injection-site reactions were more frequent in the onercept groups.

Immunological effects of TNF inhibition in psoriasis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

Pharmacological inhibition of TNF-α has been shown to have beneficial effects on different pathophysiological aspects of psoriasis. Infliximab therapy in psoriasis over a period of up to 6 weeks (113) resulted in rapid and dramatic decreases in epidermal inflammation and normalization of keratinocyte differentiation in psoriatic plaques (114). As soon as 2 weeks after a single intravenous infusion of infliximab the number of epidermal CD3+ T cells had dramatically decreased in contrast to the placebo group. Importantly, the majority of epidermal T-cell depletion was observed by week 2, before the maximal clinical clearing of psoriasis observed at week 10 (114). Moreover, infliximab treatment significantly reduced epidermal thickness in the psoriatic plaque, which positively correlated with the decrease in epidermal T cells. An additional local effect observed in the skin of psoriatic patients was the conversion of K16+ epidermal keratinocytes to K16 keratinocytes after infliximab treatment. This therapeutic effect, which was observed in all samples tested after 10 weeks of therapy represents a normalization of epidermal differentiation. Furthermore, keratinocyte ICAM-1 expression – which is known to be induced by TNF-α and IFN-γin vitro (133) and which can only be found in psoriatic plaques and not in non-lesional skin – was halted after 10 weeks of infliximab treatment. Importantly, the decrease in the number of epidermal T cells, epidermal thickness, and normalization of K16 keratin and ICAM-1 expression all correlated with the PASI and physicians global assessment (PGA) clinical efficacy measures (114). The rapid nature and completeness of this response could be explained by in vitro data demonstrating that infliximab, bound to transmembrane TNF-α, can induce antibody-dependent, cell-mediated, or complement-mediated cytotoxicity (134). If these in vitro data can be confirmed in infliximab-treated psoriatic plaques, they could possibly provide a mechanism for rapid elimination of activated T cells and antigen-presenting cells. Another mechanism of action for infliximab that has been demonstrated in CD is the induction of apoptosis of monocytes that could account for the monocytopenia that is commonly observed after infliximab treatment and could also account for the anti-inflammatory properties (135). Activated gut T lymphocytes produce less IFN-γ after infliximab treatment (136). Interestingly, in patients with rheumatoid arthritis, infliximab increases CD4+ and CD8+ T-lymphocyte counts after one day of treatment followed by a marked decrease in monocyte counts after 7 days. Additionally, in this patient population, serum concentrations of IL-1, IL-6, and soluble CD14 were significantly diminished after TNF-blockade with infliximab (137). Another very intriguing mechanistic concept that is being discussed for infliximab's potent and broad anti-inflammatory effects is the blockade of NF-κB. This transcription factor plays a pivotal role in the pathogenesis of many inflammatory diseases such as inflammatory bowel disease (134), and recent evolving evidence also gives it a central role in the pathogenesis of inflammatory skin disease (138) Indeed, the down-regulation of intracellular gene expression as controlled by NF-κB has been demonstrated in CD patients under infliximab treatment (139) and the observation of ICAM-1 down-regulation in psoriasis under infliximab therapy fits this concept, because this adhesion molecule is among the many inflammatory molecules that are transcriptionally regulated by NF-κB (140).

Also, etanercept modulates several biological systems that are influenced by TNF-α. These include the expression of a variety of adhesion molecules that regulate leukocyte migration, serum levels of proinflammatory cytokines such as IL-1, IL-6, IL-8, IL-10, and IL-12, and neutral metalloproteinases (141). Other systemic anti-inflammatory effects observed after etanercept treatment were the decrease in the median erythrocyte sedimentation rate from 22 to 5 and also the decrease of the median levels of C-reactive protein from 14 at baseline to 4 after 12 weeks of treatment whereas these parameters were unchanged in the placebo group (124). Interestingly, effective therapy with etanercept in generalized pustular exacerbation of psoriasis after withdrawal of cyclosporin-A induced circulating IL-10 levels while decreasing serum levels of IL-6 and IL-8 during treatment (132).

Clinical effects of anti-TNF therapy in other dermatological disorders

  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

Pyoderma gangrenosum

A small study investigated the efficacy of infliximab treatment in three patients suffering from inflammatory bowel disease (IBD)-associated peristomal pyoderma gangrenosum (PPG). All patients tolerated the drug without significant side effects. While two patients with PPG recovered completely following the administration of infliximab, one patient had a partial response to the drug (142).

Subcorneal pustular dermatosis

Infliximab has also been shown to be highly effective in the treatment of recalcitrant subcorneal pustular dermatosis (Sneddon–Wilkinson disease) (143). However, the patient developed a suberythrodermia in spite of infliximab therapy.

Hidradenitis suppurativa

Hidradenitis suppurativa (HS) is found associated with CD. A dramatic improvement after treatment with infliximab was achieved for both refractory fistulizing CD and axillary HS (144).

Graft-vs.-host disease

Acute graft-vs.-host disease (aGvHD) is a serious complication of allogeneic peripheral blood stem cell transplantation (PBSCT). Four patients with severe aGvHD refractory to steroids and severe intestinal involvement in addition to skin and/or liver disease were treated with infliximab (10 mg/kg once a week until clinical improvement). In three of four patients, a complete resolution of diarrhea and significant improvement of skin and liver disease were observed (145).


A patient with idiopathic panniculitis of the lower legs refractory to steroids received an induction regimen of infliximab 3 mg/kg at weeks 0, 2, and 6 and responded to therapy with improved general well being and significant reductions in arthralgia and bone pain. A repeat biopsy showed slight hyperkeratosis without significant inflammation. The patient was also able to taper steroid use (146).

Cicatricial pemphigoid

A patient with long-standing cicatricial pemphigoid recalcitrant to established treatment regimens responded rapidly and lastingly to therapy etanercept (147).

Langerhans' cell histiocytosis

Langerhans' cell histiocytosis is an accumulation of dendritic Langerhans' cells and granulomatous lesions in various organs of unknown cause. An elevation in the expression of the inflammatory cytokine TNF-α has been reported in lesional T cells in Langerhans' cell histiocytosis, and treatment with etanercept normalized fever and sedimentation rate in a young patient with the disease and led to the regression of the spleen and thymus involvement of the disease (148).

Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors

  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

Cytokines did not evolve as mediators of disease

Prototypes of cytokines such as IL-1 or TNF-α and their respective signaling mechanisms can be found in insects; indeed, NF-κB and the signal transducing domains of the mammalian IL-1 receptor are found in Drosophila (149). With the use of anti-TNF-based therapies for the treatment of psoriasis, there is little question that TNF-α plays an important role in this disease. However, proinflammatory cytokines such as IL-1 and TNF-α did not evolve to harm the host but rather evolved as beneficial to the host. Other cytokines such as IL-18 and IL-15, which also play a role in autoimmune diseases, similarly evolved as beneficial to the host, particularly in terms of host defense to infection. The use of anticytokine therapies has, in fact, revealed the importance of some cytokines to host defense in humans. The level of risk to the host, however, remains controversial because of absence of required data: 1) what is the risk of infection in patients with psoriasis treated with immunosuppressive agents such as MTX without an anti-TNF-α-based therapies; 2) how many patient years of therapy are used to calculate the denominator; and 3) what is the true number of the numerator when reporting infection is voluntary? In the case of infliximab, a single injection of the antibody is carried forward and added to the database for number of patient years of treatment whereas this is not the method used to calculate the database for anakinra or etanercept.

In general, a host challenged with an infection musters the assistance of these cytokines in order to contain the microbe from spreading and to eliminate the infection by microbicidal mechanisms. For non-intracellular bacteria, the effector cell is the phagocyte, most notably the neutrophil. These cells have the machinery to engulf the organism and kill it in the phagolysosome. Gram-positive bacteria such as streptococci and staphylococci are examples of neutrophil and cell adhesion molecule-dependent host defense. In the case of macrophage intracellular microorganisms, the killing mechanisms of the infected macrophage involves IFN-γ and NO production. However, obligate intracellular organisms such as Mycobacterium tuberculosis, Listeria, Histoplasma, and Salmonella can survive and replicate inside the macrophage by usurping killing. Most notably, cytokines such as TNF-α and IFN-γ increase the killing mechanisms, and a balance is achieved whereby a microorganism such as M. tuberculosis can survive inside a macrophage, but further spread and overt disease is kept to a minimum by a competent immune system. Reducing TNF-α activity impairs the host's mechanisms of killing intracellular organisms by reducing the synergism of TNF-α plus IFN-γ, which regulate NO production.

Cytokines act at several levels in this assistance to combat Gram-positive infection: inducing the expression of endothelial and cellular adhesion molecules to facilitate the adherence of phagocytes to surfaces is an important first step. Next, increasing the production of local chemokines results in emigration of the phagocyte into the infected tissue. By stimulating NO production, cytokines provide the chemical agents for killing the microbe. Cytokines such as IL-1 and TNF-α are expressed during this time of burgeoning infection; however, production of cytokines is tightly regulated such that when the infection is cleared, production of these proinflammatory cytokines stops. Thus, the rapid ‘on’ and rapid ‘off’ production limits any negative effect of cytokine-mediated inflammation. In the case of a chronic disease, the activator arm of the cytokine cascade stays in the ‘on’ mode. As a result, there are increasingly greater amounts of the cytokines produced day after day and year after year. It appears that in diseases such as psoriasis, the production of these cytokines is dysregulated by the disease processes.

For the patient, the aim of therapy is to block the inflammatory and keratinocyte proliferation effects of these cytokines. For a disease such as psoriasis, the therapeutic effect is a reduction in cellular migration into the skin. Although decreased cell adhesion, cellular infiltration and a reduction in the inflammatory mediators such as NO reduce host defense, the net effect of anticytokine therapy appears to benefit the host. This may be due to the fact that simple pharmacokinetics limit the duration of depressed host function and allow the host's mechanisms for fighting infection to be transiently suppressed in somewhat the same result as once a day or every other day corticosteroids accomplish. One benefit of shorter acting agents such as soluble TNF receptors is that host-defense functions related to TNF return to the normal range more frequently compared to long-acting antibodies to TNF-α.

Infections in patients treated with TNF blockers

Since the introduction of therapies using TNF blockers, increased infections have not been unexpected. Of these, risk of serious infections with Gram-positive organisms is increased compared to Gram-negative infections. Voluntary reports of bacterial infections, however, do not reflect the true incidence because diagnosis and treatment are fairly routine. In contrast, because of their uncommon occurrence, infections with intracellular organisms are increasingly reported and analyzed. Other complications such as the development of lupus-like disease, demyelinating disease, and worsening of heart failure were unexpected and, in fact, contrary to the concept of reducing TNF activity (150). On March 4, 2003, the US FDA convened the Arthritis Advisory Board in order to review complications of TNF blockers (inert FDA web site for March 4th meeting). Examination of the briefing documents from this meeting reveals that patient years of etanercept usage is associated with fewer reports of opportunistic infections, including tuberculosis, than patient years of infliximab or adalimumab. Moreover, in calculating patient years of infliximab usage, a single administration of this antibody is carried forward indefinitely to the total years of use thus enlarging the denominator and reducing the number of cases per patient years of exposure. Host defenses against intracellular organisms such as Listeria (151), Histoplasma (152), Pneumocystis, or Salmonella (153) dominate with the use of infliximab. In the case of infections with M. tuberculosis, 50% are extrapulmonary or disseminated.

The two anti-TNF-α monoclonal antibodies used to treat rheumatoid arthritis are of the IgG1 class and revealed an increase in the number of opportunistic infections. The incidence of infections in rheumatoid arthritis is increased compared to the population without the disease and is thought to be due to the immunosuppressive component of the disease in the absence of treatment with agents such as prednisone, methotrexate, or cyclosporine. In rheumatoid arthritis patients treated with these agents, infections are increased further. Studies on infections in patients with plaque psoriasis are less comprehensive, but one can assume that the use of methotrexate or cyclosporine in this disease also increases the risk of infections. However, there are few reports of opportunistic infections in patients with plaque psoriasis even with treatment with methotrexate or cyclosporine.

In the case of M. tuberculosis in rheumatoid arthritis, more than 50% of the reported cases are of extrapulmonary infections and disseminated forms. Most are reactivation disease and in the case of the fully human anti-TNF-α adalimumab, the effect appears to be dose-dependent (154). Etanercept-associated reactivation of M. tuberculosis voluntary reports are lower than those of the two monoclonal antibodies, but in those cases 50% are the disseminated disease (155). As discussed below, treatment with etanercept is also associated with increased opportunistic fungal infections, but there are lower number of voluntary reports in the postmarketing period based on number of patient years of exposure. Opportunistic infections are similar to those that occur in patients treated with immunosuppressive drugs but are also observed in human immunodeficiency virus 1 (HIV)-1-infected persons with CD4+ T cells below 200/ml. In both examples, the immunosuppression is global in that the production as well as the activity of several cytokines are reduced by immunosuppressive drugs such as cyclosporine and corticosteroids and the loss of CD4+ T cells. It is remarkable that the specific blockade of a single cytokine (TNF-α) results in the same portfolio of infections. There are, however, some notably exceptions; in the case of TNF blocker therapies, there has been few cases of oral candidiasis or reactivation of cytomegalovirus infections. Herpes zoster infections, although common in patients with autoimmune diseases has not increased greatly as seen with HIV-1 infections.

Cell death by IgG1 antibodies expressing membrane TNF-α and the importance of membrane TNF-α for impaired host-defense mechanisms

The full understanding of host-defense and cytokine includes an understanding of reversible and irreversible mechanisms of action. Soluble receptors to TNF and any monoclonal antibody to TNF-α will neutralize TNF-α activity present in the extracellular space regardless of its affinity or IgG class type or whether soluble type I or type II TNF receptor. However, a great deal of TNF-α in disease is present as an integral transmembrane protein; ‘membrane’ TNF-α is fully active as described above. In fact, mice with a specific deletion in the human TNF-α molecule for cleavage and secretion of TNF-α develop severe ankylosing arthritis by 6 weeks of age (156). In these mice, there is no TNF-α in the extracellular space and all TNF-α is present as membrane TNF-α. In addition, these mice require both the type I as well as the type II TNF receptors for activity (21). Therefore, ‘membrane’ TNF-α is active by virtue of cell–cell contact. This mechanism is often called ‘juxtacrine’. In fact, a great deal of cytokine-mediated autoimmune and inflammatory disease is the result of cell–cell juxtacrine contact activation rather than soluble cytokines in the extracellular space. Disease in mice due to membrane TNF-α is treatable with either soluble TNF receptors or monoclonal anti-TNF-α antibodies. Indeed, in humans, a great deal of the disease modification by TNF blockers is likely due to prevention of membrane TNF-α activities. In the skin, contact between activated CD4+ T cells and macrophages as well as with dendritic cells contribute to the pathogenesis of disease. In general, CD4+ T cells as well as CD8+ T cells express primarily membrane TNF-α whereas macrophages and dendritic cells secrete TNF-α.

As discussed above, the ‘off’ rate of TNF-α binding to the type II cell-bound receptor is rapid when compared to the binding to the type I receptor. In many ways, the same type of rapid ‘off’ rate is observed when one considers membrane TNF-α and the binding of the type II soluble receptor. The soluble TNF receptor type II as the Fc bivalent construct (etanercept) binds to membrane TNF-α, but this interaction is reversible in that the inhibition of biological activity is less than that of infliximab and the binding is of a lower affinity (157). In contrast, infliximab forms more stable complexes with membrane TNF-α relative to the complexes formed with etanercept. Moreover, a greater number of infliximab molecules bound to membrane TNF-α with higher avidity compared to etanercept. This was supported by the observation that infliximab exhibited a greater reduction in the biological activity of membrane TNF-α-bearing cells on endothelial target cells compared to etanercept (157). These studies are consistent with the concept that the action of etanercept on membrane as well as soluble TNF-α is ‘reversible,’ whereas the binding of monoclonal anti-TNF-α antibodies is more effective for either soluble TNF-α in the extracellular space or membrane TNF-α.

In vitro, cells expressing membrane TNF-α are vulnerable to death by either complement-mediated cell lysis or antibody-dependent cytotoxicity when exposed to anti-TNF-α (infliximab) (134,157). The mechanism of cell death is due to cross-linking of membrane TNF-α by infliximab and activation of the so-called ‘membrane attack complex’ by activated complement. Cell death in vitro is likely facilitated by the addition of exogenous complement during the assay conditions. However, clinical studies reveal a different mechanism of cell death associated with infliximab. Four hours following an infusion of infliximab into patients with CD, circulating blood monocytes undergo death by caspase-dependent apoptosis (135). In peripheral blood, anti-TNF-α monoclonal antibodies of the IgG1 class kill CD4+ T cells by caspase-dependent apoptosis (158). Unlike HIV-1 infection, these IgG1 antibodies target CD4+ T cells, CD8+ T cells, macrophages, dendritic and natural killer cells, or any cell wherever TNF-α is expressed on their respective membranes but spare cells without membrane-surface TNF-α.

In contrast, etanercept does not lyse cells expressing TNF-α on the membrane (157). There is no activation of caspase-3 in CD4+ T cells exposed to etanercept (158,159). Although etanercept is constructed with the complement receptor domains of human IgG1, the hinge region of the fusion of the Fc chain to the p75 extracellular domain of the TNF receptor is missing one of the CH2 groups and hence the structure is predicted to be more rigid than the Fc of natural antibodies. Although it remains unclear whether the deleted CH2 group explains the difference between etanercept and either infliximab or adalimumab, etanercept does not activate complement and does not lyse cells expressing membrane TNF-α. Soluble TNF receptors lenercept, onercept, and pegsunercept also do not activate complement. Thus, decreased host-defense associated with any method of TNF-α neutralization (infliximab, adalimumab, anti-TNF-α monoclonal antibodies of the IgG4 class, etanercept, lenercept, onercept, and pegsunercept) should be contrasted with the additional role of cell death of membrane TNF-α cells by IgG1 monoclonal antibodies. In the latter case, cell death can be viewed as an ‘irreversible’ event with the progressive loss of CD4+ and CD8+ T cells. In the case of soluble receptors to TNF or anti-TNF-α of the IgG4 class, the reduction in TNF-α activity by cell–cell contact is ‘reversible’.

Anti-TNF-α monoclonal antibodies affect the number of peripheral CD4+ and CD8+ T cells

The vast majority of studies on the effects of anti-TNF-α monoclonal antibodies on T cells are performed in cells from patients with CD. Gastroenterologists readily perform colonic biopsies to monitor the severity of disease, and hence tissues are available for clinical and immunological analyses. In contrast, rheumatologists in the US rarely perform synovial biopsies to assess disease progression. However, in patients with AS, studies on the percent of circulating CD4+ and CD8+ T cells have reported significant changes following a course of either infliximab or etanercept. Using specific reagents that detect cells producing TNF-α or IFN-γ, Zou and colleagues remove PBMCs before and after therapeutic treatment of AS patients with either infliximab or etanercept compared to placebo-treated patients (160,161). For infliximab, the assessment is 6 and 12 weeks after two consecutive infusions and for etanercept, after 12 weeks of bi-weekly injections. Although the clinical efficacy for both TNF blockers is similar and statistically significant, the effect of each TNF blocker is different.

PBMCs were removed after 6 weeks from both placebo- and infliximab-treated AS patients and stimulated in vitro with the combination of phorbol myristate acetate/ionomycin for non-specific increases in cytokines or the G1 domain of aggrecan as a specific antigen stimulant. The cells were subjected to fluorescence-activated cell sorter analysis using T-cell surface markers of CD4 and CD8. In addition, staining for intracellular IFN-γ and TNF-α was performed. Endotoxin was used to stimulate the monocytes in the PBMC preparations, and the amount of TNF-α and IL-10 was assessed by specific ELISA. After 12 weeks, AS patients infused with infliximab exhibited a highly significant reduction in the number of CD4+ T cells and CD8+ T cells expressing IFN-γ or TNF-α (160). In fact, the reduction in these cells had already reached statistical significance after 6 weeks whereas there were no changes in the placebo-treated patients. After 6 weeks, placebo-treated AS patients were infused with infliximab and a similar significant decrease in IFN-γ- and TNF-α-expressing CD4+ T cells and CD8+ T cells was observed (160). Similar results were observed in PBMCs stimulated with the specific antigen. Unexpectedly, there was no reduction in the amount of TNF-α produced from endotoxin-stimulated PBMCs in the patients treated with infliximab.

Using an identical experimental protocol, 10 patients with AS were treated with 25 mg etanercept twice weekly and 10 AS patients received bi-weekly injections of placebo (161). In contrast to the studies reported above with infliximab, 12 weeks of etanercept resulted in an increase in the percentage of CD4+ and CD8+ T cells expressing TNF-α and IFN-γ (161). The increases were highly significant, and there were no changes in the placebo-treated group of AS patients. Similar increases were also noted in PBMC stimulated with the specific antigen of aggrecan-derived peptides. These studies provide the first evidence that similar to patients with CD, AS patients representing a rheumatological disease also exhibit a reduction in CD4+ T cells when treated with infliximab. The increase in the percentage of CD4+ and CD8+ T cells expressing TNF-α or IFN-γ supports the concept that neutralization of TNF-αper se does not result in a reduction in circulating CD4+ T cells. Loss of CD4+ T cells expressing IFN-γ may contribute to the high incidence of granulomatous disease reported in patients receiving infliximab (154).

Neutralization of peripheral TNF-α does not induce a down-regulation of the ability of T cells to produce TNF-α but rather an up-regulation, possibly due to a counter-regulatory mechanism.

Role of IFN-γ in infections associated with TNF blockers

One can conclude from large population studies that a well-contained granuloma containing live M. tuberculosis is not incompatible with long-life as long as the patient's immune system is fully functional. A fully functional immune system includes cytokine production, particularly IFN-γ. Most notably, mutations in any of five genes which control IFN-γ production result in severe and life-threatening mycobacterial disease from birth (162). These include the IFN-γ receptor type I and type II, the p40 chain of IL-12, the IL-12 receptor β1, and the intracellular transcription factor known as STAT 1. These human studies are supported by a large body of animal studies showing failure to contain M. tuberculosis infection in mice treated with antibodies to or lacking production of TNF-α, IL-18, IL-12, or IFN-γ. In addition to the disseminated nature of M. tuberculosis infection, the organisms are similar to those observed in HIV-1-infected patients with CD4+ T-cell counts less than 200/µl. There are, however, notable exceptions; in HIV-1 infection, disseminated infection by M. avium intracellulare is common whereas this is not a common infection in patients receiving TNF blockers. Also, oral candidiasis is not seen in rheumatoid arthritis treated with TNF blockers but occurs often in the low CD4+ T cell HIV-1 population. It is important to note that patients treated with anti-TNF-α monoclonal antibodies of the IgG1 class do not exhibit low number of circulating CD4+ T cells. In fact, functional characteristics of peripheral T cells in rheumatoid arthritis patients treated with anti-TNF-α monoclonal antibodies exhibit improved function, a general observation of nearly all anti-inflammatory drugs in this disease.

Nevertheless, in both the HIV-1 population and the rheumatoid arthritis population receiving anti-TNF-α monoclonal antibodies of the IgG1 class, the target cell is a CD4+ T cell. In the case of HIV-1 infection, any cell expressing CD4 on its surface together with CCR5 (the chemokine receptor 5) can be infected and dies an apoptotic death upon viral production. Thus HIV-1 spares the CD8+ T cells while progressively killing off the CD4+ T cell population. In the case of anti-TNF-α monoclonal antibodies of the IgG1 class, any cell expressing TNF-α on its membrane is vulnerable to cell death by either complement-mediated cell lysis or antibody-dependent cytotoxicity (134,157). Recent studies demonstrate that following an infusion of infliximab into patients with CD, circulating blood monocytes undergo death by caspase-dependent apoptosis (135). In peripheral blood, anti-TNF-αmonoclonal antibodies of the IgG1 class kill CD4+ T cells by caspase-dependent apoptosis (158). Unlike HIV-1 infection, these IgG1 antibodies target CD4+ T cells, CD8+ T cells, macrophages, dendritic and natural killer cells, or any cell wherever TNF-α is expressed on their respective membranes but spare cells without membrane-surface TNF-α.

In contrast, etanercept does not lyse cells or activate caspase-3 in cells expressing membrane TNF-α (158). Although etanercept is constructed with the complement receptor domains of human IgG1, the hinge region of the fusion of the Fc chain to the p75 extracellular domain of the TNF receptor is missing one of the CH2 groups, and hence the structure is predicted to be more rigid than the Fc of natural antibodies. Whether this is the explanation or not, etanercept does not activate complement and does not lyse cells expressing membrane TNF-α. It is also the case of other soluble receptors of TNF receptors (lenercept, onercept, and pegsunercept). Thus, one must consider any clinical event indicative of decreased host defense against infection associated with therapy using any method of TNF-α neutralization (infliximab, adalimumab, anti-TNF-α monoclonal antibodies of the IgG4 class, etanercept, lenercept, onercept, and pegsunercept) with infection associated with monoclonal antibodies that fix complement and also result in loss of membrane TNF-α (infliximab and adalimumab).

Testing for risk of reactivation of M. tuberculosis

It is good medical practice to assess the risk of reactivation of tuberculosis before initiating anti-TNF-α therapy. Because routine M. tuberculosis testing before initiating TNF blockers was instituted, the number of cases with reactivation M. tuberculosis infection has decreased substantially. However, most physicians do not assess anergy when testing for response to purified protein derivative. In fact, in patients with rheumatoid arthritis, the incidence in anergy has been demonstrated. In a study of recall antigens in 48 newly diagnosed patients with rheumatoid arthritis prior to any treatment with DMARDs, 43.75% of the patients exhibited anergy compared to 2% of healthy age- and sex-matched controls (163). The loss of skin test responses was not related to clinical or biochemical markers of disease severity. Similar to nearly all anti-inflammatory treatment regimens, depression of cell-mediated immunity can be partially reversed, which was observed in this study using methotrexate but not cyclosporine or hydroxychloroquine (163). Thus, there are no guarantees when it comes to predictions of reactivation of infection with M. tuberculosis.

Although screening for M. tuberculosis infection is now routine and opportunistic infections can be treated with appropriate antimicrobials, one is far less sanguine about an increased risk of non-Hodgkin's lymphoma (B-cell lymphoma, NHL) (164).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References

These recent results indicate that inhibition of TNF-α could become a highly promising option in the treatment of psoriasis and psoriatic arthritis. However, severe infections in some of the patients treated with etanercept, infliximab, and humira underline the need for careful control until further experience with this treatment has been accumulated. It is likely that anti-TNF strategies will lead to registration of new drugs for the treatment of psoriasis and to supplementation of existing therapeutic options, which represents considerable progress from 20 years ago when cytokines were first discovered. These approaches are, as a majority of novel immunological therapies currently in early clinical development, biologicals – recombinant proteins and antibodies. Although these molecules have offered great relief for the psoriasis patient, they do have obvious drawbacks. Biologicals must be administered by injection, which is quite inconvenient for the patients; they are expensive in many cases, and the induction of neutralizing antibodies has to be excluded for each product. There is already emerging evidence that many of these treatments generally do not achieve complete remission in all patients. Therefore, it would be interesting to identify which patients are suitable for each specific anti-TNF therapy, perhaps by using pharmacogenomic techniques that are sure to be described in the future. The crucial question remains, however, how such an identification of certain subpopulation of patients can be achieved. Perhaps analyses of cytokine receptor expression or polymorphisms may be helpful in this process. Only if this can be achieved, immunotherapies might conceivably be tailored to suit individual patients. Moreover, it will definitively have to be established which combinations of anti-TNF strategies with various other established and experimental approaches will yield synergistic effects in the patient. Finally, the results of clinical trials are significantly contributing to our further understanding of the disease, indicating that TNF is a crucial cytokine in psoriasis. This in turn will generate momentum for further better-targeted pharmacological action in the future. Hence, receptor blockade, inhibition of TNF signaling downstream from the receptor, and inhibition of TNF formation are certainly interesting options, which are indeed already pursued by several pharmaceutical companies. This approach for example includes the inhibition of the enzymes p38 MAP kinase, PDE4, and TACE, or the inhibition of kinases in TNF-signaling like I kappa B Kinase (IKK). These molecules can be targeted with small molecules, which might allow for the development of orally available drugs in the future (Fig. 8).


Figure 8. Cellular and intracellular targets of anti-tumor necrosis factor (anti-TNF)-α drug development [adapted from (166)]. A large number of agents, including both biologic agents and small molecule compounds targeting TNF activity, are in early stages of development. Apart from the biologics discussed in more depth in this review, most of the agents shown here have been investigated in experimental disease models but lack clinical safety and efficacy data. Protherics Molecular Design and Faulding & Co. have developed CytoTAb, an anti-TNF antibody fragment preparation generated by immunizing sheep against TNF. PassTNF (Verigen and Italchimici) is a porcine polyclonal antibody. Advanced biotherapy is testing AGT-1, a mixture of three antibodies raised against TNF, interferon-α (IFN-α), and IFN-γ. Thalidomide enhances the degradation of messanger RNA encoding for TNF. CDC-801, CDC-501 (Celgene), FRP-33, and (R)-FPTN (Ishihara Sangyo) are Thalidomide analogs. CDC-801 also inhibits PDE4 (phosphodiesterase 4). Scio-469 (Scios Inc.) and VX-745 (Vertex Pharmaceuticals) are p38 MAP kinase inhibitors that are able to modulate TNF, interleukin-1, and cycooxygense-2 (COX-2). Solimastat is a matrix metalloproteinase (MMP) inhibitor. Several companies are developing inhibitors of the metalloproteinase TNF-α-converting enzyme (TACE) that would prevent the release of soluble TNF receptors into the circulation. Among the kinase inhibitors in development is a molecule directed against I kappa B kinase 2 (IKK2; Celgene). RDP-58 (SangStat Medical) is a decepeptide that inhibits TNF expression at the translational level. NFκB, nuclear factor kappa B; mRNA, messenger RNA; PDE, phosphodiesterase.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Immunobiology of TNF-α
  5. Expression of TNF in psoriasis
  6. Strategies for the inhibition of TNF
  7. Clinical effects of anti-TNF therapy in psoriasis
  8. Immunological effects of TNF inhibition in psoriasis
  9. Clinical effects of anti-TNF therapy in other dermatological disorders
  10. Differences in host-defense impairment between anti-TNF-α antibodies and soluble TNF receptors
  11. Perspectives
  12. Acknowledgements
  13. References
  • 1
    Abrams J R, Kelley S L, Hayes E et al. Blockade of T lymphocyte costimulation with cytotoxic T lymphocyte-associated antigen 4-immunoglobulin (CTLA4Ig) reverses the cellular pathology of psoriatic plaques, including the activation of keratinocytes, dendritic cells, and endothelial cells. J Exp Med 2000: 192: 681694.
  • 2
    Abrams J R, Lebwohl M G, Guzzo C A et al. CTLA41g mediated blockade of T-cell costimulation in patients with psoriasis vulgaris. J Clin Invest 1999: 103: 12431252.
  • 3
    Asadullah K, Docke W D, Ebeling M et al. Interleukin 10 treatment of psoriasis: clinical results of a phase 2 trial. Arch Dermatol 1999: 135: 187192.
  • 4
    Asadullah K, Volk H D, Sterry W. Novel immunotherapies for psoriasis. Trends Immunol 2002: 23: 4753.
  • 5
    Baker B S, Fry L. The immunology of psoriasis. Br J Dermatol 1992: 126: 19.
  • 6
    Haranaka K, Carswell E A, Williamson B D, Prendergast J S, Satomi N, Old L J. Purification, characterization, and antitumor activity of nonrecombinant mouse tumor necrosis factor. Proc Natl Acad Sci USA 1986: 83: 39493953.
  • 7
    Aggarwal B B, Aiyer R A, Pennica D, Gray P W, Groeddel D V. Human tumour necrosis factor: structure and receptor interactions. Ciba Found Symp 1987: 131: 3951.
  • 8
    Pennica D, Hayflick J S, Bringman T S, Palladino M A, Goeddel D V. Cloning and expression in Escherichia coli of the cDNA for murine tumor necrosis factor. Proc Natl Acad Sci USA 1985: 82: 60606064.
  • 9
    Beutler B. Tumor Necrosis Factors. The Molecules and Their Emerging Role in Medicine. New York, NY: Raven Press, 1992.
  • 10
    Kriegler M, Perez C, DeFay K, Albert I, Lu S D. A novel form of TNF/cachectin is a cell surface cytotoxic transmembrane protein: ramifications for the complex physiology of TNF. immune responses in MHC class II-deficient mice. Cell 1988: 53: 4553.
  • 11
    Moss M L, Jin S L C, Milla M E et al. Cloning of a disintegrin metalloproteinase that processes precursor tumor-necrosis factor-alpha. Nature 1997: 385: 733736.
  • 12
    Peschon J J, Slack J L, Reddy P et al. An essential role for ectodomain shedding in mammalian development. Science 1998: 282: 12811284.
  • 13
    Bazzoni F, Beutler B. The tumor necrosis factor ligand and receptor families. N Engl J Med 1996: 334: 17171725.
  • 14
    Beutler B, Brown T. Polymorphism of the mouse TNF-alpha locus: sequence studies of the 3′-untranslated region and first intron. Gene 1993: 129: 279283.
  • 15
    Han J, Brown T, Beutler B. Endotoxin-responsive sequences control cachectin/tumor necrosis factor biosynthesis at the translational level. J Exp Med 1990: 171: 465475.
  • 16
    Piecyk M, Wax S, Beck A R et al. TIA-1 is a translational silencer that selectively regulates the expression of TNF-alpha. EMBO J 2000: 19 (15): 41544163.
  • 17
    Kontoyiannis D, Pasparakis M, Pizarro T T, Cominelli F, Kollias G. Impaired on/off regulation of TNF biosynthesis in mice lacking TNF AU-rich elements: implications for joint and gut-associated immunopathologies. Immunity 1999: 10: 387398.
  • 18
    Schottelius A J, Mayo M W, Sartor R B, Baldwin A S J. Interleukin-10 signaling blocks inhibitor of kappaB kinase activity and nuclear factor kappaB DNA binding. J Biol Chem 1999: 274 (45): 3186831874.
  • 19
    Wong B, Choi Y. Pathways leading to cell death in T cells. Curr Opin Immunol 1997: 9: 358364.
  • 20
    Butler D M, Malfait A, Mason L J et al. DBA/1 mice expressing the human TNF-alpha transgene develop a severe, erosive arthritis. J Immunol 1997: 159 (6): 28672876.
  • 21
    Alexopoulou L, Pasparakis M, Kollias G. A murine transmembrane tumor necrosis factor transgene induces arthritis by cooperative p55/p75 tumor necrosis factor receptor signaling. Eur J Immunol 1997: 27 (10): 25882592.
  • 22
    C K, Chlipala E S, Dinarello C A, Reznikov L L, Moldawer L L, Bendele A M. Clinical and histopathologic characterization of arthritis in male and female tumor necrosis factor-alpha knockout (TNF-alpha–/–) and membrane-bound TNF-alpha transgenic (TNF-alphaTgA86) mice injected with mycoplasma pulmonies or mycoplasma arthritidis. Arthritis Rheum 1999: 42: S120.
  • 23
    Marinova-Mutafchieva L, Williams R O, Masoon L J Mauri C, Feldmann M, Maini R N. Dynamics of proinflammatory cytokine expression in the joints o mice with collagen-induced arthritis (CIA). Clin Exp Immunol 1997: 107: 507512.
  • 24
    Mussener A, Litton M J, Lindros E, Klareskog L. Cytokine production in synovial tissue of mice with collagen-induced arthritis (CIA). Clin Exp Immunol 1997: 107: 485493.
  • 25
    Jeng K C G, Liu M T, Lan J L, Wu C.H, Wong D W, Cheung B M. H. Collagen induces cytokine production by synovial fluid mononuclear cells in rheumatoid arthritis. Immune responses in MHC class II-deficient mice. Immunol Lett 1995: 45: 1317.
  • 26
    Thorbecke G J, Shah R, Leu C.H, Kuruvilla A P, Hardison A M Palladino M A. Involvement of endogenous tumor necrosis factor alpha and transforming growth factor beta during induction of collagen type II arthritis in mice. Proc Natl Acad Sci USA 1992: 89: 73757379.
  • 27
    Williams R O, Feldmann M, Maini R N. Anti-tumor necrosis factor ameliorates joint disease in murine collagen-induced arthritis. Proc Natl Acad Sci USA 1992: 89: 97849788.
  • 28
    Piguet P F, Grau G E, Vesin C, Loetscher H, Gentz R, Lesslauer W. Evolution of collagen arthritis in mice is arrested by treatment with anti-tumour necrosis factor (TNF) antibody or a recombinant soluble TNF receptor. Immunology 1992: 77: 510514.
  • 29
    Probert L, Plows D, Kontogeorgos G, Kollias G. The type I interleukin-1 receptor acts in series with tumor necrosis factor (TNF) to induce arthritis in TNF-transgenic mice. Eur J Immunol 1995: 25 (6): 17941797.
  • 30
    Ulfgren A K, Andersson U, Engstrom M, Klareskog L, Maini R N, Taylor P C. Systemic anti-tumor necrosis factor alpha therapy in rheumatoid arthritis down-regulates synovial tumor necrosis factor alpha synthesis. Arthritis Rheum 2000: 43 (11): 23912396.
  • 31
    Van Den Berg W B, Joosten L A, Kollias G, Van De Loo F A. Role of tumour necrosis factor alpha in experimental arthritis: separate activity of interleukin 1beta in chronicity and cartilage destruction. Ann Rheum Dis 1999: 58 (Suppl. 1): I40I48.
  • 32
    Van Den Berg W B. Uncoupling of inflammatory and destructive mechanisms in arthritis. Semin Arthritis Rheum 2001: 30 (5): 717.
  • 33
    Kuiper S, Joosten L A B, Bendele A M et al. Different roles of tumour necrosis factor a and interleukin 1 in murine streptococcal cell wall arthritis. Cytokine 1998: 10 (9): 690702.
  • 34
    Joosten L A, Helsen M M A, Saxne T, Van-de-Loo F A J, Heinegard D, Van-den-Berg W B. IL-1alphabeta blockade prevents cartilage and bone destruction in murine type II collagen-induced arthritis, whereas TNF-alpha blockade only ameliorates joint inflammation. J Immunol 1999: 163 (9): 50495055.
  • 35
    Joosten L A, Helsen M M A, Vandeloo F A J, Vandenberg W B. Anticytokine Treatment of Established Type I I, Collagen-Induced Arthritis In D B A/1 Mice: a comparative study using anti-TNF-alpha, anti-IL-1-alpha/beta, and IL−1ra. Immune responses in MHC class II-deficient mice. Arthritis Rheum 1996: 39 (5): 797809.
  • 36
    Campbell I K, O'Donnell K, Lawlor K E, Wicks I P. Severe inflammatory arthritis and lymphadenopathy in the absence of TNF. J Clin Invest 2001: 107 (12): 15191527.
  • 37
    Van Der Poll T, Van Deventer S J, Hack C E et al. Effects of leukocytes after injection of tumor necrosis factor into healthy humans. Blood 1992: 79: 693698.
  • 38
    Van Der Poll T, Van Deventer S J, Buller H R, Sturk A, Ten Cate J W. Comparison of the early dynamics of systemic prostacyclin release after administration of tumor necrosis factor and endotoxin to healthy humans. J Infect Dis 1991: 164: 599601.
  • 39
    Van Der Poll T, Romijn J A, Endert E, Borm J J, Buller H R, Sauerwein H P. Tumor necrosis factor mimics the metabolic response to acute infection in health humans. Am J Physiol 1991: 261: E457E465.
  • 40
    Van Der Poll T, Levi M, Buller H R et al. Fibrinolytic response to tumor necrosis factor in healthy subjects. J Exp Med 1991: 174: 729732.
  • 41
    Spriggs D R, Sherman M L, Michie H et al. Recombinant human tumor necrosis factor administered as a 24-hour intravenous infusion. A phase I and pharmacologic study. J Natl Cancer Inst 1988: 80 (13): 10391044.
  • 42
    Zamkoff K W, Newman N B, Rudolph A R, Young J, Poiesz B J. A phase I trial of subcutaneously administered recombination tumor necrosis factor to patients with advanced malignancy. J Biol Response Mod 1989: 8 (5): 539552.
  • 43
    Aderka D, Sorkine P, Abu-Abeid S et al. Shedding kinetics of soluble tumor necrosis factor (TNF) receptors after systemic TNF leaking during isolated limb perfusion. Relevance to the pathophysiology of septic shock. J Clin Invest 1998: 101: 650659.
  • 44
    Zwaveling J H, Maring J K, Clarke F L et al. High plasma tumor necrosis factor (TNF)-alpha concentrations and a sepsis–like syndrome in patients undergoing hyperthermic isolated limb perfusion with recombinant TNF-alpha, interferon-gamma, and melphalan. Crit Care Med 1996: 24: 765770.
  • 45
    Edwards C K. III PEGylated recombinant human soluble tumour necrosis factor receptor type I (r-Hu-sTNF-RI) novel high affinity TNF receptor designed for chronic inflammatory diseases. Ann Rheum Dis 1999: 58 (Suppl. 1): 173181.
  • 46
    Bemelmans M H, Gouma D J, Buurman W A. Tissue distribution and clearance of soluble murine TNF receptors in mice. Cytokine 1994: 6: 608615.
  • 47
    Seckinger P, Isaaz S, Dayer J M. Purification and biologic characterization of a specific tumor necrosis factor alpha inhibitor. J Biol Chem 1989: 264: 1196611973.
  • 48
    Seckinger P, Isaaz S, Dayer J M. A human inhibitor of tumor necrosis factor alpha. J Exp Med 1988: 167 (4): 15111516.
  • 49
    Seckinger P, Zhang J H, Hauptmann B, Dayer J M. Characterization of a tumor necrosis factor alpha (TNF-alpha) inhibitor: evidence of immunological cross-reactivity with the TNF receptor. Proc Natl Acad Sci USA 1990: 87: 51885192.
  • 50
    Frishman J I, Edwards C K, Sonnenberg M G, Kohno T, Cohen A M, Dinarello C A. Tumor necrosis factor (TNF)-alpha-induced interleukin-8 in human blood cultures discriminates neutralization by the p55 and p75 TNF soluble receptors. J Infect Dis 2000: 182 (6): 17221730.
  • 51
    Grell M, Wajant H, Zimmerman G, Scheurich P. The type I receptor (CD120a) is the high-affinity receptor for soluble tumor necrosis factor. Proc Natl Acad Sci USA 1998: 95 (2): 570575.
  • 52
    Tartaglia L A, Pennica D, Goeddel D V. Ligand passing: the 75-kd tumor necrosis factor (TNF) receptor recrutits TNF for signaling by the 55 Kd TNF receptor. J Biol Chem 1993: 268 (25): 1854218548.
  • 53
    Grell M, Zimmerman G, Hulser D, Pfizenmaier K, Scheurich P. TNF receptors TR60 and TR80 can mediate apoptosis via induction of distinct signal pathways. J Immunol 1994: 153: 19631972.
  • 54
    Weiss T, Grell M, Siemienski K et al. TNFR80-dependent enhancement of TNFR60-induced cell death is mediated by TNFR-associated factor 2 and is specific for TNFR60. J Immunol 1998: 161: 31363142.
  • 55
    Malinin N L, Boldin M P, Kovalenko A V, Wallach D. MAP3K-related kinase involved in NK-κΒ induction by TNF, CD95 and IL-1. Nature 1997: 385: 540544.
  • 56
    Baeuerle P A, Baltimore D. N F-Kappa-B: ten years after. Cell 1996: 87: 1320.
  • 57
    Wang C Y, Mayo M W, Baldwin A S J. TNF-and cancer therapy-induced apoptosis: potentiation by inhibition of NF-kappaB. Science 1996: 274 (5288): 784787.
  • 58
    Pimentel-Muinos F X, Seed B. Regulated commitment of TNF receptor signaling: a molecular switch for death or activation. Immunity 1999: 11 (6): 783793.
  • 59
    Peschon J J, Torrance D S, Stocking K L et al. TNF receptor-deficient mice reveal divergent roles for p55 and p75 in several models of inflammation. J Immunol 1998: 160: 943952.
  • 60
    Nowak M, Gaines G C, Rosenberg J et al. LPS-induced liver injury in d-galactosamine-sensitized mice requires secreted TNF-alpha and the TNF-p55 receptor. Am J Physiol Regul Integr Comp Physiol 2000: 278 (5): R1202R1209.
  • 61
    Welborn M B, Vanzee K, Edwards P D et al. A Human Tumor Necrosis Factor P75 receptor agonist stimulates in vitro T cell proliferation but does not produce inflammation or shock in the baboon. J Exp Med 1996: 184: 165171.
  • 62
    Van Zee K J, Stackpole S A, Montegut W J et al. A human tumor necrosis factor (TNF) alpha mutant that binds exclusively to the p55 TNF receptor produces toxicity in the baboon. J Exp Med 1994: 179: 11851191.
  • 63
    Sheehan K C, Pinckard J K, Arthur C D, Dehner L P, Goeddel D V, Schreiber D. Monoclonal antibodies specific for murine p55 and p75 tumor necrosis receptors: identification of a novel in vivo role for p75. J Exp Med 1995: 181: 607617.
  • 64
    Pfeffer K, Matsuyama T, Kuendig T M et al. Mice deficient for the 55 kd tumor necrosis factor receptor are resistant to endotoxic shock, yet succumb to L. monocytogenes infection. Cell 1993: 73: 457467.
  • 65
    Rothe J, Lesslauer W, Loetscher H et al. Mice lacking the tumour necrosis factor receptor 1 are resistant to TNF-mediated toxicity but highly susceptible to infection by Listeria monocytogenes. Nature 1993: 364: 798802.
  • 66
    Erickson S L, De Sauvage F J, Kikly K et al. Decreased sensitivity to tumor-necrosis factor but normal T-cell development in TNF receptor-2-deficient mice. Nature 1994: 372: 560563.
  • 67
    Grell M, Douni E, Wajant H et al. The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80kDa tumor necrosis factor receptor. Cell 1995: 83: 793802.
  • 68
    Kusters S, Tiegs G, Alexopoulou L et al. In vivo evidence for a functional role of both tumor necrosis factor (TNF) receptors and transmembrane TNF in experimental hepatitis. Eur J Immunol 1997: 27 (11): 28702875.
  • 69
    Douni E, Kollias G. A critical role of the p75 tumor necrosis factor receptor (p75TNF-R) in organ inflammation independent of TNF, lymphotoxin alpha, of the p55TNF-R. J Exp Med 1998: 188 (7): 13431352.
  • 70
    Williams L M, Gibbons D L, Gearing A, Maini R N, Feldmann M, Brennan F M. Paradoxical effects of a synthetic metalloproteinase inhibitor that blocks both P55 and P75 TNF receptor shedding and TNFalpha processing in Ra synovial membrane cell cultures. J Clin Invest 1996: 97: 28332841.
  • 71
    Feldmann M. Development of anti-TNF therapy for rheumatoid arthritis. Nat Rev Immunol 2002: 2 (5): 364371.
  • 72
    Di G F, Nuki G, Duff G W. Tumour necrosis factor in synovial exudates. Ann Rheum Dis 1988: 47: 768772.
  • 73
    Feldmann M, Brennan F M, Maini R N. Role of cytokines in rheumatoid arthritis. Annu Rev Immunol 1996: 14: 397440.
  • 74
    Firestein G S, Alvaro-Garcia J M, Maki R. Quantitative analysis of cytokine gene expression in rheumatoid arthritis. J Immunol 1990: 144: 33473353.
  • 75
    Chu C Q, Field M, Feldmann M, Maini R N. Localization of tumor necrosis factor alpha in synovial tissues and at the cartilage-pannus junction in patients with rheumatoid arthritis. Arthritis Rheum 1991: 34 (9): 1125.
  • 76
    Alvaro-Gracia J M, Zvaifler N J, Brown C B, Kaushansky K, Firestein G S. Cytokines in chronic inflammatory arthritis. VI. Analysis of the synovial cells involved in granulocyte-macrophage colony-stimulating factor production and gene expression in rheumatoid arthritis and its regulation by IL-1 and tumor necrosis factor-alpha. J Immunol 1991: 146 (10): 33653371.
  • 77
    Alvaro-Gracia J M, Zvaifler N J, Firestein G S. Cytokines in chronic inflammatory arthritis. V. mutual antagonism between interferon-gamma and tumor necrosis factor-alpha on HLA-DR expression, proliferation, collagenase production, and granulocyte macrophage colony-stimulating factor production by rheumatoid arthritis synoviocytes. J Clin Invest 1990: 86 (6): 17901798.
  • 78
    Alvaro-Gracia J M, Zvaifler N J, Firestein G S. Cytokines in chronic inflammatory arthritis. V. Mutual antagonism between interferon-gamma and tumor necrosis factor-alpha on HLA-DR expression, proliferation, collagenase production, and granulocyte macrophage colony-stimulating factor production by rheumatoid arthritis synoviocytes. J Clin Invest 1990: 86 (6): 17901798.
  • 79
    Alsalameh S, Winter K, Al-Ward R, Wendler J, Kalden J R, Kinne R W. Distribution of TNF-alpha, TNF-R55 and TNF-R75 in the rheumatoid synovial membrane: TNF receptors are localized preferentially in the lining layer; TNF-alpha is distributed mainly in the vicinity of TNF receptors in the deeper layers. Scand J Immunol 1999: 49 (3): 278285.
  • 80
    Robak T, Gladalska A, Stepien H. The tumor necrosis factor family of receptors/ligands in the serum of patients with rheumatoid arthritis. Eur Cytokine Netw 1998: 9: 145154.
  • 81
    Steiner G, Studnicka-Benke A, Witzmann G, Hofler E, Smolen J. Soluble receptors for tumor necrosis factor and interleukin-2 in serum and synovial fluid of patients with rheumatoid arthritis, reactive arthritis and osteoarthritis. J Rheumatol 1995: 22 (3): 406412.
  • 82
    Saklatvala J. Tumour necrosis factor alpha stimulates resorption and inhibits synthesis of proteoglycan in cartilage. Nature 1986: 322: 547549.
  • 83
    Kristensen M, Chu C Q, Eedy D J, Feldmann M, Brennan F M, Breathnach S M. Localization of tumour necrosis factor-alpha (TNF-alpha) and its receptors in normal and psoriatic skin: epidermal cells express the 55-kD but not the 75-kD TNF receptor. Clin Exp Immunol 1993: 94: 354362.
  • 84
    Beyaert R, Schulze-Osthoff K, Van Roy F, Fiers W. Synergistic induction of interleukin-6 by tumor necrosis factor and lithium chloride in mice: possible role in the triggering and exacerbation of psoriasis by lithium treatment. Eur J Immunol 1992: 22 (8): 21812184.
  • 85
    Bonifati C, Carducci M, Cordiali F P et al. Correlated increases of tumour necrosis factor-alpha, interleukin-6 and granulocyte monocyte-colony stimulating factor levels in suction blister fluids and sera of psoriatic patients – relationships with disease severity. Clin Exp Dermatol 1994: 19 (5): 383387.
  • 86
    Ettehadi P, Greaves M W, Wallach D, Aderka D, Camp R D. Elevated tumour necrosis factor-alpha (TNF-alpha) biological activity in psoriatic skin lesions. Clin Exp Immunol 1994: 96 (1): 146151.
  • 87
    Olaniran A K, Baker B S, Paige D G, Garioch J J, Powles A V, Fry L. Cytokine expression in psoriatic skin lesions during PUVA therapy. Arch Dermatol Res 1996: 288 (8): 421425.
  • 88
    Nickoloff B J, Karabin G D, Barker J N et al. Cellular localization of interleukin-8 and its inducer, tumor necrosis factor-alpha in psoriasis. Am J Pathol 1991: 138 (1): 129140.
  • 89
    Uyemura K, Yamamura M, Fivenson D F, Modlin R L, Nickoloff B J. The cytokine network in lesional and lesion-free psoriatic skin is characterized by a T-helper type 1 cell-mediated response. J Invest Dermatol 1993: 101 (5): 701705.
  • 90
    Asadullah K, Prosch S, Audring H et al. A high prevalence of cytomegalovirus antigenaemia in patients with moderate to severe chronic plaque psoriasis: an association with systemic tumour necrosis factor alpha overexpression. Br J Dermatol 1999: 141 (1): 94102.
  • 91
    Ritchlin C, Haas-Smith S A, Hicks D, Cappuccio J, Osterland C K, Looney R J. Patterns of cytokine production in psoriatic synovium. J Rheumatol 1998: 25 (8): 15441552.
  • 92
    Partsch G, Steiner G, Leeb B F, Dunky A, Broll H, Smolen J S. Highly increased levels of tumor necrosis factor-alpha and other proinflammatory cytokines in psoriatic arthritis synovial fluid. J Rheumatol 1997: 24: 518523.
  • 93
    Suda T, Takahashi T, Golstein P, Nagata S. Molecular cloning and expression of the Fas ligand, a novel member of the tumor necrosis factor family. Cell 1993: 75 (6): 11691178.
  • 94
    Moreland L W, McCabe D P, Caldwell J R et al. Phase I/II trial of recombinant methionyl human tumor necrosis factor binding protein PEGylated dimer in patients with active refractory rheumatoid arthritis. J Rheumatol 2000: 27 (3): 601609.
  • 95
    Moreland L W, Baumgartner S W, Schiff M H et al. Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. New Engl J Medical 1997: 337 (3): 141147.
  • 96
    Moreland L W, Heck J, Koopman W J. Biologic agents for treating rheumatoid arthritis: concepts and progress. Arthritis Rheum 1997: 40: 397409.
  • 97
    Jacobi A, Manger B, Schuler G, Hertl M. Therapeutic application of TNF-alpha inhibitors infliximab and etanercept in inflammatory skin disorders. J Deutschen Dermatologischen Gesellschaft 2003: 1 (4): 259272.
  • 98
    Maini R, St Clair E W, Breedveld F et al. Infliximab (chimeric anti-tumor necrosis factor α monoclonal antibody) versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. Lancet 1999: 354: 19321939.
  • 99
    Maini R N, Elliott M, Brennan F M, Williams R O, Feldmann M. TNF blockade in rheumatoid arthritis: implications for therapy and pathogenesis. APMIS 1997: 105 (4): 257263.
  • 100
    Santora L C, Krull I S, Grant K. Characterization of recombinant human monoclonal tissue necrosis factor-alpha antibody using cation-exchange HPLC and capillary isoelectric focusing. Anal Biochem 1999: 275 (1): 98108.
  • 101
    Kempeni J. Update on D2E7: a fully human anti-tumour necrosis factor alpha monoclonal antibody. Ann Rheum Dis 2000: 59 (Suppl 1): 4445.
  • 102
    Christen U, Thuerkauf R, Stevens R, Lesslauer W. Immune response to a recombinant human TNFR55-IgG1 fusion protein: auto- antibodies in rheumatoid arthritis (RA) and multiple sclerosis (MS) patients have neither neutralizing nor agonist activities. Hum Immunol 1999: 60 (9): 774790.
  • 103
    Furst D, Weisman M, Paulus H et al. Neutralization of TNF by Lenercept. (TNFr55-IgG1, Ro 45–2081) in patients with rheumatoid arthritis treated for 3 months; results of an US phase II trial. Arthritis Rheum 1996: 39: S243.
  • 104
    Sander O, Rau R, Van Riel P L et al. (TNFR55-IgG1, Ro 45–2081) in patients with rheumatoid arthritis treated for 3 months: results of a European phase II trial. Arthritis Rheum 1996: 39: S242.
  • 105
    Sander O, Rau R. Long term observation of 80 patients treated with TNF alpha receptor-fusion protein (TNFR55- IgG1, Ro 45–2081). Arthritis Rheum 2000: 40 (9): S224.
  • 106
    Kneer J, Luedin E, Lesslauer W, Birnboeck H, Stevens R M. An assessment of the effect of anti-drug antibody formation of the pharmacokinetics and pharmacodynamics of sTNFr55-IgG (Ro 45-2081-Lenercept [LEN]) (S.C.) in patients with rheumatoid arthritis. Arthritis Rheum 1998: 41: S58.
  • 107
    Solorzano C C, Ksontini R, Pruitt J H et al. Involvement of 26-kDa cell associated TNFa in experimental hepatitis and exacerbation of liver injury with a matrix metalloproteinase inhibitor. J Immunol 1997: 158 (1): 414419.
  • 108
    Solorzano C C, Kaibara A, Hess P J et al. Pharmacokinetics, immunogenicity and efficacy of dimeric TNFR binding proteins in healthy and bacteremic baboon. J Appl Physiol 1998: 84 (4): 11191130.
  • 109
    Serono. Positive results of onercept in psoriasis and psoriatic arthritis in phase II trials. 9th International Psoriasis Symposium in New York June 2003.
  • 110
    Serono. Serono announces positive results of onercept in psoriasis and psoriatic arthritis in phase II trials. 6-22-2003. (
  • 111
    Oh C J, Das K M, Gottlieb A B. Treatment with anti-tumor necrosis factor α (TNF-α) monoclonal antibody dramatically decreases the clinical activity of psoriasis lesions. J Am Acad Dermatol 2000: 42: 829830.
  • 112
    Kirby B, Marsland A M, Carmichael A J, Griffiths C E. Successful treatment of severe recalcitrant psoriasis with combination infliximab and methotrexate. Clin Exp Dermatol 2001: 26 (1): 2729.
  • 113
    Chaudhari U, Romano P, Mulcahy L D, Dooley L T, Baker D G, Gottlieb A B. Efficacy and safety of infliximab monotherapy for plaque-type psoriasis: a randomised trial. Lancet 2001: 357 (9271): 18421847.
  • 114
    Gottlieb A B, Masud S, Ramamurthi R et al. Pharmacodynamic and pharmacokinetic response to anti-tumor necrosis factor-alpha monoclonal antibody (infliximab) treatment of moderate to severe psoriasis vulgaris. J Am Acad Dermatol 2003: 48 (1): 6875.
  • 115
    Gottlieb A B, Li S, Evans R, Menter A. Infliximab in the treatment of psoriasis: results from the first 10 weeks of the phase II trial. 61th Annu Meet Am Acad Dermatol (March 21–26, 2003 San Francisco).
  • 116
    Buchholz C. Promising new results reported for Remicade (infliximab) in the treatment of psoriasis. 3-1-2003. (
  • 117
    Gulliver WP, Brown-Maher T, Tabrizi M. Low dose infliximab improves psoriasis. 61th Annu Meet Am Acad Dermatol (March 21–26, 2003 San Francisco).
  • 118
    Newland M R, Weinstein A, Kerdel F. Rapid response to infliximab in severe pustular psoriasis, von Zumbusch type. Int J Dermatol 2002: 41 (7): 449452.
  • 119
    O'Quinn R P, Miller J L. The effectiveness of tumor necrosis factor alpha antibody (Infliximab) in treating recalcitrant psoriasis: a report of 2 cases. Arch Dermatol 2002: 138 (5): 644648.
  • 120
    Cauza E, Spak M, Cauza K, Hanusch-Enserer U, Dunky A, Wagner E. Treatment of psoriatic arthritis and psoriasis vulgaris with the tumor necrosis factor inhibitor infliximab. Rheumatol Int 2002: 22 (6): 227232.
  • 121
    Antoni C, Dechant C, Hanns-Martin Lorenz P D et al. Open-label study of infliximab treatment for psoriatic arthritis: clinical and magnetic resonance imaging measurements of reduction of inflammation. Arthritis Rheum 2002: 47 (5): 506512.
  • 122
    Ogilvie A L, Antoni C, Dechant C et al. Treatment of psoriatic arthritis with antitumour necrosis factor-alpha antibody clears skin lesions of psoriasis resistant to treatment with methotrexate. Br J Dermatol 2001: 144 (3): 587589.
  • 123
    Antoni C, Smolen J. The Infliximab Multinational Psoriatic Arthritis Controlled Trial (IMPACT). 61th Annu Meet J Am Acad Dermatol (March 21–26, 2003 San Francisco).
  • 124
    Mease P J, Goffe B S, Metz J, VanderStoep A, Finck B, Burge D J. Etanercept in the treatment of psoriatic arthritis and psoriasis: a randomised trial. Lancet 2000: 356 (9227): 385390.
  • 125
    Mease P. Psoriatic arthritis: the role of TNF inhibition and the effect of its inhibition with etanercept. Clin Exp Rheumatol 2002: 20 (Suppl 28): S116.
  • 126
    Mease P J. Etanercept in psoriatic arthritis. 59th Annu Meet Am Acad Dermatol. Washington, DC 2001.
  • 127
    Gottlieb A B, Goffe B S, Burge D J, Zitnik R, Wayne T. Etanercept inhibits radiographic progression in patients with psoriatic arthritis. J Investig Dermatol 2003: 120 (7): 430.
  • 128
    Leonardi C, Powell J I, Matheson R. T et al. Etanercept as monotherapy in patients with psoriasis. N Engl J Med 2003: 349: 20142022.
  • 129
    Leonardi C, Gottlieb A B, Zitnik R. Efficacy and safety of Enbrel in patients with psoriasis: results of a phase 3 study. J Investig Dermatol 2003: 120 (7): 430.
  • 130
    Leonardi C L, Powers J L, Matheson R T et al. Etanercept as monotherapy in patients with psoriasis. N Engl J Med 2003: 349 (21): 20142022.
  • 131
    Gottlieb A B, Feng A, Zitnik R. Prolonged response durability following etanercept monotherapy. J Invest Dermatol 2003: 120 (7): 430.
  • 132
    Kamarashev J, Lor P, Forster A, Heinzerling L, Burg G, Nestle F O. Generalised pustular psoriasis induced by cyclosporin a withdrawal responding to the tumour necrosis factor alpha inhibitor etanercept. Dermatology 2002: 205 (2): 213216.
  • 133
    Norris D A. Cytokine modulation of adhesion molecules in the regulation of immunologic cytotoxicity of epidermal targets. J Invest Dermatol 1990: 95 (6 Suppl.): 111S120S.
  • 134
    Scallon B J, Moore M A, Trinh H, Knight D M, Ghrayeb J. Chimeric anti-TNF-alpha monoclonal antibody cA2 binds recombinant transmembrane TNF-alpha and activates immune effector functions. Cytokine 1995: 7 (3): 251259.
  • 135
    Lugering A, Schmidt M, Lugering N, Pauels H G, Domschke W, Kucharzik T. Infliximab induces apoptosis in monocytes from patients with chronic active Crohn's disease by using a caspase-dependent pathway. Gastroenterology 2001: 121 (5): 11451157.
  • 136
    Agnholt J, Kaltoft K. Infliximab downregulates interferon-gamma production in activated gut T-lymphocytes from patients with Crohn's disease. Cytokine 2001: 15 (4): 212222.
  • 137
    Lorenz H M, Antoni C, Valerius T et al. In vivo blockade of TNF-alpha by intravenous infusion of a chimeric monoclonal TNF-alpha antibody in patients with rheumatoid arthritis. Short term cellular and molecular effects. J Immunol 1996: 156: 16461653.
  • 138
    Bell S, Degitz K, Quirling M, Jilg N, Page S, Brand K. Involvement of NF-[kappa]B signalling in skin physiology and disease. Cellular Signalling 2003: 15 (1): 17.
  • 139
    Nikolaus S, Raedler A, Kuhbacker T, Sfikas N, Folsch U R, Schreiber S. Mechanisms in failure of infliximab for Crohn's disease. Lancet 2000: 356 (9240): 14751479.
  • 140
    Baldwin A S Jr. The NF-kappa B and I kappa B proteins: new discoveries and insights. Annu Rev Immunol 1996: 14: 649683.
  • 141
    Immunex Inc. Enbrel, prescribing information. 11-1-1998.
  • 142
    Mimouni D, Anhalt G J, Kouba D J, Nousari H C. Infliximab for peristomal pyoderma gangrenosum. Br J Dermatol 2003: 148 (4): 813816.
  • 143
    Voigtlander C, Luftl M, Schuler G, Hertl M. Infliximab (anti-tumor necrosis factor alpha antibody): a novel, highly effective treatment of recalcitrant subcorneal pustular dermatosis (Sneddon–Wilkinson disease). Arch Dermatol 2001: 137 (12): 15711574.
  • 144
    Sullivan T P, Welsh E, Kerdel F A, Burdick A E, Kirsner R S. Infliximab for hidradenitis suppurativa. Br J Dermatol 2003: 149 (5): 10461049.
  • 145
    Kobbe G, Schneider P, Rohr U et al. Treatment of severe steroid refractory acute graft-versus-host disease with infliximab, a chimeric human/mouse antiTNFalpha antibody. Bone Marrow Transplant 2001: 28 (1): 4749.
  • 146
    Humeniuk J M, Gliptis E. Infliximab in the treatment of panniculitis. World Congress Of Dermatology Paris July 1–5 2002.
  • 147
    Sacher C, Rubbert A, Konig C, Scharffetter-Kochanek K, Krieg T, Hunzelmann N. Treatment of recalcitrant cicatricial pemphigoid with the tumor necrosis factor alpha antagonist etanercept. J Am Acad Dermatol 2002: 46 (1): 113115.
  • 148
    Henter J I, Karlen J, Calming U, Bernstrand C, Andersson U, Fadeel B. Successful treatment of Langerhans′-cell histiocytosis with etanercept. N Engl J Med 2001: 345 (21): 15771578.
  • 149
    Gay N J, Keith F J, Drosophila T. IL-1 receptor. Nature 1991: 351: 355356.
  • 150
    Bozkurt B, Torre A G, Warren M S et al. Results of targeted anti-tumor necrosis factor therapy with etanercept (ENBREL) in patients with advanced heart failure. Circulation 2001: 103 (8): 10441047.
  • 151
    Slifman N R, Gershon S K, Lee J H, Edwards E T, Braun M M. Listeria monocytogenes infection as a complication of treatment with tumor necrosis factor alpha-neutralizing agents. Arthritis Rheum 2003: 48 (2): 319324.
  • 152
    Lee J H, Slifman N R, Gershon S K et al. Life-threatening histoplasmosis complicating immunotherapy with tumor necrosis factor alpha antagonists infliximab and etanercept. Arthritis Rheum 2002: 46 (10): 25652570.
  • 153
    Netea M G, Radstake T, Joosten L A, Van Der Meer J W, Barrera P, Kullberg B J. Salmonella septicemia in rheumatoid arthritis patients receiving anti-tumor necrosis factor therapy: association with decreased interferon-gamma production and Toll-like receptor 4 expression. Arthritis Rheum 2003: 48 (7): 18531857.
  • 154
    Keane J, Gershon S, Wise R P et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med 2001: 345 (15): 10981104.
  • 155
    Manadan A M, Mohan A K, Cote R T, Siegel J N, Sequiera W, Block J A. Tuberculosis and etanercept treatment. Arthritis Rheum 2002: 46 (Suppl): S166.
  • 156
    Georgopoulos S, Plows D, Kollias G. Transmembrane TNF is sufficient to induce localized tissue toxicity and chronic inflammatory arthritis in transgenic mice. J Inflamm 1996: 46 (2): 8697.
  • 157
    Scallon B, Cai A, Solowski N et al. Binding and functional comparisons of two types of tumor necrosis factor antagonists. J Pharmacol Exp Ther 2002: 301: 418426.
  • 158
    Van den Brande J M, Braat H, Van Den Brink G R et al. Infliximab but not etanercept induces apoptosis in lamina propria T-lymphocytes from patients with Crohn's disease. Gastroenterology 2003: 124 (7): 17741785.
  • 159
    Van Deventer S J. Transmembrane TNF-alpha, induction of apoptosis, and the efficacy of TNF-targeting therapies in Crohn's disease. Gastroenterology 2001: 121 (5): 12421246.
  • 160
    Zou J, Rudwaleit M, Brandt J, Thiel A, Braun J, Sieper J. Down-regulation of the nonspecific and antigen-specific T cell cytokine response in ankylosing spondylitis during treatment with infliximab. Arthritis Rheum 2003: 48 (3): 780790.
  • 161
    Zou J, Rudwaleit M, Brandt J, Thiel A, Braun J, Sieper J. Up regulation of the production of tumour necrosis factor alpha and interferon gamma by T cells in ankylosing spondylitis during treatment with etanercept. Ann Rheum Dis 2003: 62 (6): 561564.
  • 162
    Ottenhoff T H, Verreck F A, Lichtenauer-Kaligis E G, Hoeve M A, Sanal O, Van Dissel J T. Genetics, cytokines and human infectious disease: lessons from weakly pathogenic mycobacteria and salmonellae. Nat Genet 2002: 32 (1): 97105.
  • 163
    Coaccioli S, Di Cato L, Marioli D et al. Impaired cutaneous cell-mediated immunity in newly diagnosed rheumatoid arthritis. Panminerva Med 2000: 42 (4): 263266.
  • 164
    Brown S L, Greene M H, Gershon S K, Edwards E T, Braun M M. Tumor necrosis factor antagonist therapy and lymphoma development: twenty-six cases reported to the Food and Drug Administration. Arthritis Rheum 2002: 46 (12): 31513158.
  • 165
    C K, Borcherding S, Zhang J, Borcherding D R, Laskin D L., Edwards III The role of tumor necrosis factor-alpha (TNF-alpha) in acute and chronic inflammatory responses: novel therapeutic approaches. In: SchookL B, ed. Xenobiotic Induced Inflammation: Roles of Cytokines and Growth Factors. New York, NY: Academic Press, 1994: 97147.
  • 166
    Decision Resources I. TNF-alpha inhibitors (Immune, Inflammatory Disorders Study # 34). Waltham, MA, Decision Resources, Inc. Immune and Inflammatory Disorders. 2001.
  • 167
    Aggarwal B B. Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol 2003: 3 (9): 745756.
  • 168
    Wanke L A, Burge D J. Etanercept improves health-related quality of life in patients with psoriatic arthritis. 60th Annu Meet Am Acad Dermatol (February 22–27, New Orleans) 2002.
  • 169
    Lebwohl M, Gottlieb A B, Mease P J, Burge D J. Etanercept improves psoriasis activity in patients with psoriatic arthritis: results of a phase 3 multicenter clinical trial. 60th Annu Meet Am Acad Dermatol (February 22–27, New Orleans) 2003.
  • 170
    Gottlieb A B, Metheson R T, Lowe N, Zitnik R. Efficacy of Enbrel in patients with psoriasis. J Invest Dermatol 2002: 119: 234.