TNF-α is a ligand that binds the TNF receptor (TNF-R) mediating inflammatory and immunomodulatory activity. Two isoforms of TNF-R exist: the p55 isoform (TNFr1) and the p75 isoform (TNFr2). TNF-Rs are found as either membrane-bound or soluble forms. Five TNF-α inhibitors are commercially available: infliximab, adalimumab, etanercept, golimumab, and certolizumab. The first three have been studied clinically for the treatment of uveitis and related ocular inflammatory processes. Each medication inhibits various portions of the TNF-α pathway, but each differs in its molecular structure and/or method of TNF-inhibition (Figure 1).
Infliximab is a chimeric human-murine monoclonal antibody with human constant IgG1κ region and murine variable sequences, which binds to soluble and cell membrane surface-bound TNF-α, thereby inhibiting its action. Adalimumab is a fully human monoclonal antibody to TNF-α and is thought to have less immunogenicity than infliximab due to fewer murine protein sequences in the structure of the monoclonal antibody. Etanercept is the soluble p75 TNF-receptor, which prevents TNF-α and β from binding to extracellular TNF-R. Although etanercept initially met some success for treatment of uveitis in small, retrospective series, comparative retrospective studies have suggested that etanercept may be less efficacious than the monoclonal antibodies in decreasing the frequency of uveitis exacerbations.
The majority of reports in the literature on TNF-α inhibition involve clinical experience with infliximab for the primary treatment of uveitis or as a corticosteroid-sparing agent. The efficacy of infliximab, which was initially developed as treatment for rheumatoid arthritis, has been reported for other inflammatory conditions including adult and pediatric Crohn disease, ulcerative colitis, HLA-B27–associated arthritis, psoriasis, and, juvenile idiopathic arthritis (JIA).
Inflammatory ocular conditions with reported response to infliximab include rheumatoid arthritis–associated scleritis and peripheral ulcerative keratitis, Behçet disease, intermediate uveitis, retinal vasculitis, and sarcoidosis. Currently, few prospective studies using TNF-α inhibition are available.
In one prospective, phase II, open-label study of infliximab for refractory autoimmune uveitis including pars planitis, sarcoidosis, Crohn disease, and Behçet disease, 78% of patients with uveitis responded to infliximab. Patients were administered 3 or 5 mg/kg intravenous infliximab infusions at weeks 0, 2, and 6 and assessed for a positive response at the 10-week follow-up. Clinical criteria for success were met if patients achieved one or more of the following outcomes: 2-step reduction in inflammation (reduction in anterior chamber cell or vitreous haze as defined by Nussenblatt criteria), 50% reduction in systemic immunosuppressive medication, decrease in inflammatory signs by optical coherence tomography or fluorescein angiography, and vision improvement of at least 2 lines. In this prospective series, 24 (78%) of 31 patients achieved initial clinical criteria for success. At the 1-year follow-up, 15 (50%) of 31 patients maintained clinical improvement. Adverse effects reported in this initial trial included vitreous hemorrhage, pulmonary embolus, drug-related lupus-like syndrome, and congestive heart failure.
One particularly promising area for the use of TNF-α blockers appears to be in the treatment of the ocular manifestations of Behçet disease, including panuveitis, retinal vasculitis, and secondary complications including cystoid macular edema. Limited prospective data and retrospective series have suggested infliximab to be of value for long-term therapy, with complete clinical remission reported in up to 75% of patients with Behçet disease–associated uveitis over a 12-month follow-up period. However, if infliximab therapy is withheld, most patients develop recurrent disease activity. In one study in which patients were administered infliximab in combination with conventional immunosuppressive medication during an infusion period (weeks 0 through 22) and conventional immunosuppressive alone during an observation period (weeks 23 through 53), 12 of 13 patients treated experienced one or more uveitis exacerbations during the observation period while the frequency of attacks was significantly lower during the infusion period. The inability to taper off infliximab therapy once it has been initiated without inflammatory recurrence raises important questions for the long-term management of these patients.
The use of infliximab for scleritis has been reported, particularly in patients with systemic conditions such as rheumatoid arthritis. One small prospective pilot study demonstrated a 2-step decrease in scleral inflammation (National Institutes of Health grading scheme) in several patients treated with infliximab (5 mg/kg) at baseline, week 2, week 6, and every 4 weeks thereafter. Retrospective evidence has also demonstrated the efficacy of infliximab for some patients with scleritis refractory to other immunomodulatory medications.
Infliximab has also been used in the pediatric population for several autoimmune conditions including JIA and inflammatory bowel disease. Its use for JIA-associated uveitis and other types of pediatric uveitis has been limited, but results of small, retrospective series have demonstrated efficacy for the acute reduction of active uveitis, as well as its use as a corticosteroid-sparing agent. In one study, 5 of 6 patients with pediatric uveitis refractory to other therapies (including idiopathic uveitis, JIA-associated uveitis, idiopathic retinal vasculitis, and pars planitis) were able to taper off their corticosteroids with initial doses of infliximab at 5 to 10 mg/kg at 2- to 4-week intervals, followed by maintenance therapy of 5 to 18 mg/kg doses at 4- to 8-week intervals.
Another larger retrospective study of 16 pediatric patients demonstrated the efficacy of infliximab for chronic noninfectious uveitis in children with a 2-step reduction in inflammation or zero inflammation in 79% of patients at the 1-year follow-up. In addition, 69% of patients were able to discontinue topical corticosteroids. The majority of patients (15 of 16 patients) also received concomitant methotrexate. Higher doses of infliximab (10 to 20 mg/kg) have also been reported to result in rapid disease quiescence in one retrospective study. Further studies are needed to determine optimal treatment doses, induction schemes, and maintenance regimens for the pediatric population. Figure 2 illustrates the efficacy of infliximab observed in one patient with pars planitis.
Dosage and Administration
Infliximab is given via an intravenous infusion, typically in 3 to 10 mg/kg doses. Concomitant administration of infliximab with methotrexate has been associated with a decrease in the formation of human anti-chimeric antibodies, which can be immunogenic and/or bind infliximab and reduce its efficacy. Higher doses of 10 to 20 mg/kg have been described in one report for the treatment of pediatric uveitis, but the optimal doses for uveitis in both pediatric and adult patients are not known.
Side Effects and Toxicity
TNF-α inhibitors have been associated with reactivation of latent tuberculosis, serious life-threatening opportunistic infections, unmasked or de novo demyelinating disease, exacerbation of congestive heart failure, abnormal auto-antibody formation (eg, anti-double stranded DNA, anti-nuclear antibodies), and lupus-like syndromes. Reactivation of herpes zoster has been reported, including thoracic zoster and herpes zoster keratitis. Optic neuritis has rarely occurred following infliximab therapy. Of concern are several recent reports of hepatosplenic T-cell lymphoma following the use of infliximab for Crohn disease in pediatric patients. Some authors have suggested a possible increased risk of lymphoma or other malignancies in rheumatoid arthritis patients treated with infliximab, which may be related to the dose and duration of treatment. However, there is an increased risk for lymphoma in rheumatoid arthritis independent of infliximab therapy, and conflicting reports exist; hence, the association of TNF blockers and lymphoma or other malignancies is unclear at this time. The toxicity of anti-TNF antibodies may be higher in children and young adults and has led to a black-box warning from the FDA about the potential increased risk of malignancy in these patient groups. Discussion of the potential long-term risks of secondary malignancy and serious infection is a necessary part of the informed consent process.
A complete medical evaluation including purified protein derivative testing, chest radiograph, routine blood chemistries, and a complete blood count should be performed when infliximab therapy is considered. In patients with suggestive neurologic findings, or those with intermediate uveitis of the pars planitis subtype, a pretreatment brain magnetic resonance imaging study is recommended to rule out demyelinating disease due to multiple sclerosis. Further laboratory testing should be tailored according to a complete review of systems and physical examination, with an experienced internist, rheumatologist, or uveitis specialist. Table 2 highlights the medical evaluation and laboratory work-up currently recommended prior to initiation of an anti-TNF agent.
Adalimumab is a fully human monoclonal IgG1 antibody, which blocks the interaction of TNF-α with the p55 and p75 TNF-α cell surface receptors. Its efficacy has been reported for rheumatoid arthritis, ankylosing spondylitis, Crohn disease, and psoriatic arthritis. Promising efficacy for uveitis has been described in several retrospective series in adult and pediatric patients.
Ophthalmic therapeutic indications for adalimumab include intermediate uveitis, JIA-associated uveitis, and Behçet disease–associated uveitis. Fewer reports are available for adalimumab than infliximab, but preliminary evidence suggests that adalimumab can be efficacious in ocular conditions in which TNF-α is a key mediator of inflammation.
Several retrospective case series have shown that adalimumab can be effective for adult Behçet disease patients. One study reported clinical improvement in 3 patients with refractory Behçet disease when they were switched from infliximab to adalimumab. Successful switching within the biologic agents has recently been reported as a therapeutic strategy for individuals in whom a recurrence develops during anti–TNF-α therapy.
Retrospective studies have demonstrated the efficacy of adalimumab for pediatric uveitis, particularly for JIA-associated uveitis. In early studies, 80% of eyes experienced reduction of inflammation, and 80% or more of patients were able to taper their concomitant topical corticosteroids with no relapses during follow-up. A more recent retrospective study evaluating 20 patients with JIA-associated chronic uveitis, using Standardization of Uveitis Nomenclature (SUN) criteria, reported improvement in uveitis activity in 7 (35%) patients and a mean reduction in the number of uveitis flares/year from 1.9 to 1.4. A prospective multicenter study is underway, with 10-week data indicating a 68% initial response, with further long-term data pending, while a larger, industry-sponsored randomized, placebo-controlled study is currently enrolling patients. Figure 3 illustrates the efficacy of adalimumab for sarcoidosis-associated panuveitis and cystoid macular edema.
Dosage and Administration
Adalimumab is given as a subcutaneous injection at a dose of 20 or 40 mg administered either once weekly or once every other week. Early phase I trials showed that weight-based dosing offered no significant pharmacokinetic advantages. Subcutaneous injections of adalimumab, although variably painful, can be administered at home and do not require visits to infusion centers.
Side Effects and Toxicity
Common side effects include injection-site reactions in up to 10% of patients receiving adalimumab. Other adverse effects identified in post-marketing surveillance of patients on adalimumab are similar to TNF-α blocker class effects as described for infliximab. Patients taking TNF-α blockers should avoid live attenuated virus vaccinations. The safety of adalimumab in pregnancy has not been determined.
Etanercept is a TNF-antagonist, which differs from the prior medications discussed, infliximab and adalimumab, in that it is a recombinant TNF-receptor fusion protein composed of the constant (Fc) portion of the human IgG1 and 2 copies of the extracellular ligand-binding portion of the TNF-α receptor p75. It also differs from the aforementioned antibodies in that it binds both TNF-α and TNF-β (also known as lymphotoxin A), but only the free cytokine, not that which is membrane-bound, and hence does not fix complement. While etanercept has been used effectively for pediatric and adult autoimmune conditions including rheumatoid arthritis, JIA, ankylosing spondylitis, and psoriatic arthritis, its efficacy for ophthalmic inflammatory disease is poor and distinctly inferior to that of infliximab.
Dosing and Administration
Etanercept is given via subcutaneous injection with a dose of 25 to 50 mg twice weekly. Its elimination half-life is slightly greater than 4 days following a single administration of 25 mg subcutaneous dose.
Initial enthusiasm for etanercept stemmed from a prospective study that demonstrated a rapid decrease in anterior chamber inflammation after the initial administration of etanercept in 63% of eyes (ie, 10 of 16). About 25% of patients demonstrated sustained remission at 1 year. These initial encouraging results were tempered, however, by a prospective masked, placebo-controlled study evaluating etanercept for JIA-associated anterior uveitis in which no significant difference was observed in the anterior chamber inflammation when comparing etanercept-treated vs placebo groups. Moreover, etanercept showed no significant efficacy over placebo in preventing uveitis relapse in patients being tapered from methotrexate in another small (n=20) randomized, controlled trial. Several retrospective studies have shown infliximab to be superior to etanercept in the control of intraocular inflammation, with greater reduction of uveitis recurrences and greater decrease in topical corticosteroid requirement.
Side Effects and Toxicity
Uveitis exacerbations in patients being treated with etanercept and the development of de novo uveitis syndromes (eg, scleritis, myositis) and optic neuritis have been reported. While these reports are concerning, larger retrospective cohort data have suggested that etanercept probably does not increase the risk of de novo uveitis. Additional retrospective evidence derived from questionnaires submitted by pediatric rheumatologists suggests that etanercept does not reduce the frequency or severity of uveitis flares in patients with previously diagnosed JIA-associated uveitis. Systemic toxicities include injection site reactions, headache, and dizziness. Serious infections and tuberculosis are rare. Given the paucity of data supporting the use of etanercept and the availability of more effective anti-TNF agents, etanercept is not regarded as first-line biologic therapy for ocular inflammation at this time.
Daclizumab is a humanized recombinant monoclonal antibody targeting Tac, a 55 kDa IL (interleukin)-2α receptor subunit expressed by most T cells, B cells, and natural killer cells. The IL-2 receptor system (IL-2R) is composed of 3 subunits (α, β, and γ), and plays a critical role in the ophthalmic immune response. Binding of daclizumab to IL-2Rα prevents formation of the high-affinity IL-2 receptor complex and leads to down-regulation of the T-cell–mediated response and antibody production. Daclizumab has been used for the prevention of renal allograft rejection and for immunosuppressive therapy following other solid organ transplantation protocols. Daclizumab is also used for the treatment of human T-cell lymphotropic virus type 1 (HTLV-1)–associated adult T-cell lymphoma/leukemia. Patients with this specific form of T-cell lymphoma/leukemia may be successfully treated with daclizumab because of the IL-2 receptor expressed on leukemic cells.
The efficacy of daclizumab for endogenous, noninfectious uveitis has been reported in several prospective trials including one multicenter trial and several retrospective case series. Conditions that have responded to daclizumab include birdshot retinochoroidopathy, sarcoidosis, intermediate uveitis, Vogt-Koyanagi-Harada disease, and scleritis. Patients with Behçet disease do not appear to derive significant benefit from daclizumab therapy.
An initial prospective pilot study demonstrated efficacy of daclizumab as a corticosteroid- and cyclosporine-sparing agent for 8 of 10 patients, with diagnoses including intermediate uveitis, sarcoidosis, Vogt-Koyanagi-Harada disease, multifocal choroiditis, and idiopathic panuveitis. A longer-term (>4 years) phase I/II study of intravenous daclizumab (phase I) followed by subcutaneous daclizumab (phase II) showed that 7 of 10 patients were able to taper off their other immunosuppressive medications completely. Interestingly, patients who were administered daclizumab with a longer dosing interval (ie, every 6-week intervals) experienced uveitis recurrences whereas those with a shorter dosing interval (ie, every 2- to 4-week intervals) did not. Pediatric patients with uveitis have also been treated with some success in limited case series.
In one prospective pilot study, a higher initial induction dose of daclizumab (8 mg/kg at week 0, 4 mg/kg at week 4, and 2 mg/kg thereafter) was efficacious in the treatment of active, noninfectious uveitis in patients unable to tolerate corticosteroids. This dosing regimen has also met some success in a prospective pilot study for JIA-associated anterior uveitis, although that study was limited by small numbers of patients and short follow-up.
Dosing and Administration
Daclizumab is typically administered via intravenous infusion (1 to 2 mg/kg every 2 to 4 weeks). It was administered at the same dose every 4 weeks in the previously described clinical trials. The subcutaneous and intravenous formulations are not currently available in the United States.
Side Effects and Toxicity
In studies of patients receiving daclizumab for renal allograft rejection, no significant difference was observed in the incidence of cancer or serious infections among patients receiving placebo vs daclizumab. However, one patient developed renal cell carcinoma while on daclizumab therapy in a previously reported National Eye Institute trial. Antibodies against the daclizumab monoclonal antibody appear in 15% of treated patients; however, the development of these antibodies does not appear to predispose patients to the development of other autoimmune toxicities and/or to dampen the efficacy of the medication.