Tumor necrosis factor α (TNF) is a pro-inflammatory cytokine produced primarily by cells of the macrophage-monocyte lineage. The biologic effects of TNF are varied, and include adhesion molecule expression, synthesis of proinflammatory cytokines, synthesis of chemokines, activation of other immune system cells (T-cells, B-cells, and macrophages), and inhibition of regulatory T-cells. TNF exists in both cell membrane-bound and soluble forms. Three agents directed against TNF are currently approved for use: infliximab (Remicade), etanercept (Enbrel), and adalimumab (Humira). Infliximab is a chimeric monoclonal antibody comprised of the human IgG1 constant region fused with the murine variable region recognizing TNF. Adalimumab has a similar structure, but is fully humanized. Infliximab and adalimumab can bind to circulating and membrane bound TNF, and can induce apoptosis in cells expressing TNF. Etanercept is a fusion protein composed of 2 extracellular p75 TNF receptor domains linked by the Fc
portion of human IgG1. Unlike infliximab and adalimumab, etanercept does not induced apoptosis in TNF-expressing cells [1
Giant cell arteritis (GCA)
Since up to 80% of patients with GCA experience complications from corticosteroid therapy, an effective adjunct therapy to allow corticosteroid reduction is needed. Small case series and a case report presented evidence that infliximab could be used as a steroid sparing agent for GCA [2
]. Based on these reports, a randomized, multicenter trial of infliximab versus placebo was conducted to determine the efficacy of infliximab in GCA [5
]. Forty-four patients newly diagnosed with GCA were randomized to receive infliximab (5 mg/kg) or placebo in a 2:1 ratio, in addition to prednisone. At 22 weeks, the proportion of patients without relapse were similar between the infliximab and placebo groups (43% versus 50% respectively, p=0.65). In addition, the proportion of patients on prednisone tapered to 10 mg/day without relapses was similar between both groups (61% for infliximab versus 75% for placebo, p=0.31). The incidence of infection was 71% in the infliximab group and 56% in the placebo group (difference of 15%, 95% CI -14-45%). With the results of this study, the authors concluded that infliximab was unlikely to have substantial efficacy in the treatment of GCA. Although there is some evidence that etanercept and adalimumab could have a role in the treatment of GCA, the data are inconclusive [6*
Takayasu's arteritis (TA)
The successful use of anti-TNF therapy for the treatment of TA has been reported by multiple investigators [8
]. The largest case series examined 25 patients with active, relapsing TA who were TA treated with infliximab (n=21) or etanercept (n=9), and followed for a median of 28 months [14*
]. Of the 9 patients initially treated with etanercept, 4 underwent complete remission and 2 experienced partial remission. Of the 6 patients who achieved remission on etanercept, 3 had disease relapses. Three patients who did not respond to etanercept were switched to infliximab and achieved complete remission.
Of the 21 patients treated with infliximab (including the 5 previously treated with etanercept), 12 achieved a complete remission and 6 achieved a partial remission. Three patients discontinued infliximab; 12 of the remaining 18 patients relapsed, and required treatment with higher doses of infliximab administered at shorter intervals.
While the positive results from these case series and case reports of the use of anti-TNF agents for refractory TA are encouraging, they need to be replicated in larger, randomized clinical trials. Currently, there is more evidence to support the use of infliximab compared to etanercept or adalimumab. One should note, however, that doses frequently had to be escalated above the baseline 3 mg/kg dose to achieve remission.
ANCA-associated vasculitis (AAV)
Little is known regarding the efficacy of anti-TNF strategies for the treatment of microscopic polyangiitis (MPA) and Churg-Strauss syndrome (CSS) [15
]. Our understanding of the potential role of these drugs for the treatment of AAV is extrapolated from our experience with Wegener's granulomatosis (WG) [18
]. The Wegener's granulomatosis Etanercept Trial (WGET) was conducted to determine if etanercept is effective for the maintenance of remission in WG. In this multicenter, placebo-controlled trial, 180 patients were randomized to receive adjunctive therapy with etanercept versus placebo, in addition to standard-of-care immunosuppression (with either cyclophosphamide or methotrexate) [20
No difference in the rates of sustained remission was observed between the etanercept and placebo groups (69.7% vs. 75.3%, p=0.39). In addition, no difference in the relative risk of disease flares was observed between the 2 groups (0.89, p=0.54). One concerning finding was the increased rate of malignancy in the etanercept group. Six solid tumors developed in patients in the etanercept group, while none were observed in the placebo group (p=0.01). Based on age- and sex-specific incidence rates from the Surveillance, Epidemiology, and End Results (SEER) database, only 1.92 solid cancers were expected in the etanercept group, leading to a standardized incidence ratio (SIR) of 3.12 (95% CI 1.15-6.80). Additional follow-up of 140 participants in the WGET showed that this increased risk in solid cancers in the etanercept group persisted 3.5 years following the conclusion of the study (SIRetanercept 4.4, SIRplacebo 1.5, p=0.01) [P. Seo, personal communication].
The negative results of the WGET were surprising. Several potential explanations exist. First, the dose of etanercept used in the WGET may not have been sufficient. Studies of etanercept in psoriasis show an increased therapeutic effect at higher doses (e.g., 50 mg subcutaneous twice weekly) [21
]. In addition, etanercept may not be effective in diseases marked by granulomatous inflammation. For example, etanercept is not effective in Crohn's disease and sarcoidosis, which are both characterized by granulomatous inflammation. In a randomized, double-blind, placebo-controlled trial of 43 patients with moderate to severe Crohn's disease, use of etanercept (25 mg subcutaneously twice weekly) was not associated with an improvement in clinical response or clinical remission rates at 2, 4, or 8 weeks when compared to placebo [23
]. In an open-label study of etanercept in stage II/III sarcoidosis, enrollment was stopped after 17 patients since the use of etanercept was associated with both early and late treatment failure [24
No randomized clinical trials have been conducted to study the efficacy of other anti-TNF agents (i.e., infliximab or adalimumab) for AAV. Four case series and a prospective open-label trial  have been published describing the use of infliximab in WG [15
]. Results of these case series cannot be combined due to the use of different treatment regimens and outcome measures. In general, the majority of patients appear to respond and enter either a complete or partial remission. However, some patients did experience flares while receiving infliximab, and serious infections were reported.
Infliximab use in Wegener's granulomatosis
The role of anti-TNF therapy for the treatment of AAV remains uncertain. Based on the WGET, etanercept should not be used to as monotherapy or adjunctive treatment to cyclophosphamide or methotrexate for induction or maintenance of remission in WG. Without randomized clinical trials, the ability of infliximab (or adalimumab) to induce or maintain remission cannot be fully assessed. Therefore, neither should not be used as first-line therapies for these vasculitides, and could be considered for refractory disease after review of the risks and benefits of therapy. Lastly, combination therapy with an anti-TNF therapy and cyclophosphamide should be used cautiously, given the increased risk of malignancy seen in the WGET and its subsequent analyses.