Abatacept for severe anti–tumor necrosis factor α refractory juvenile idiopathic arthritis–related uveitis




To evaluate the safety and efficacy of abatacept in patients with severe juvenile idiopathic arthritis (JIA)–related uveitis refractory or intolerant to immunosuppressive and anti–tumor necrosis factor α (anti-TNFα) agents.


Patients with JIA-related uveitis refractory to immunosuppressive and anti-TNFα agents were treated with intravenous abatacept (10 mg/kg monthly). Side effects, frequency of uveitis flares, and ocular complications before and after treatment were reported.


Seven patients (6 females and 1 male) with a mean uveitis duration of 11.6 years entered the study. All patients had failed previous immunosuppressive therapy and ≥2 anti-TNFα treatments. All patients responded to abatacept and 6 maintained a clinical remission after a mean of 9.2 months of treatment. One patient withdrew from the study with oral mycosis and arthritis flare; no other patients had side effects. The mean frequency of uveitis flares during the 6 months before and after treatment decreased from 3.7 to 0.7 episodes. No new ocular complications or worsening of preexisting ones were reported.


Abatacept treatment led to sustained improvement in severe anti-TNFα–resistant JIA-related uveitis and was well tolerated in all but 1 patient. These results provide new insights into a possible indication of abatacept for the treatment of uveitis.


Chronic anterior uveitis is one of the most serious manifestations of juvenile idiopathic arthritis (JIA), with the potential to cause sight-threatening ocular complications (1). Despite improvements in the treatment strategy for arthritis, less progress has been made in the treatment of uveitis, which remains one of the greatest challenges encountered by pediatric rheumatologists and ophthalmologists. Although first-line treatment for uveitis consists of topical and oral steroids with methotrexate (MTX) for more severe uveitis (2), treatment with cyclosporin A (CSA) (3), mycophenolate mofetil (MMF) (4), and, more recently, anti–tumor necrosis factor (anti-TNF) agents such as infliximab (5) and adalimumab (6), has been proposed.

Despite the effectiveness of anti-TNFα agents in chronic anterior uveitis reported in open-label studies, the percentages of success vary considerably in different case series and no controlled trials have yet been published. Side effects and poor visual outcome are still quite common in severe, refractory uveitis.

Abatacept, a selective T cell costimulation modulator, has been shown to be a valid alternative to anti-TNFα agents for the treatment of both rheumatoid arthritis and JIA, especially in refractory cases (7, 8). Furthermore, a recent case report suggested that abatacept may be effective in patients with refractory autoimmune uveitis (9).

Following from these results, we evaluated the efficacy and safety of abatacept in a series of young patients with sight-threatening, JIA-related uveitis refractory to anti-TNFα agents.


This study included JIA patients with severe bilateral chronic anterior uveitis that was refractory to both classic immunosuppressive and anti-TNFα treatments. Patients were treated with abatacept and prospectively evaluated, since treatment start, with a standardized protocol at 4 pediatric rheumatology centers. Data on uveitis activity and flare rates prior to initiation of abatacept were collected retrospectively from the patients' clinical records.

Chronic anterior uveitis was diagnosed according to the Standardization of Uveitis Nomenclature Working Group criteria, including the standardized grading schema for anterior chamber cells and anterior chamber flare (10). A uveitis flare was defined as a 2-degree increase of the level of inflammation (anterior chamber cells). Uveitis was defined as improved when a 2-degree decrease or disappearance of inflammation (anterior chamber cells) was obtained.

Structural complications of uveitis, such as cataract or cystoid macular edema, and visual acuity outcomes have also been recorded according to standardized methods (10). For assessing visual acuity, Snellen charts were used and best corrected visual acuity was reported in a decimal scale.

Ophthalmologic checks ranged from once a week to once a month, depending on the uveitis course. The arthritis course was evaluated by rheumatologic checks, performed at baseline and during followup at regular intervals of 3 months, or less if needed.

Inclusion criteria were inadequate response to standard immunosuppressive drugs (MTX, leflunomide [LEF], MMF, or CSA) and to ≥1 anti-TNF agents, with progression of ocular complications. These agents had been administered at standard dosages for several months, as follows: MTX 10–20 mg/m2 (maximum 25 mg/week); CSA 2–5 mg/kg/day (maximum 200 mg/day); MMF 0.5–1 gm/m2/day (maximum 2 gm/day); LEF 100 mg for 3 days, and 20 mg daily thereafter; etanercept 0.4 mg/kg twice weekly (maximum 50 mg/week); infliximab 3–5 mg/kg for 4–6 weeks (maximum 350 mg); and adalimumab 20–40 mg/m2 every 2 weeks (maximum 40 mg). After a 4-week washout period, abatacept was administered at a dosage of 10 mg/kg (up to a maximum of 750 mg) by 30-minute intravenous (IV) infusion at weeks 0, 2, and 4, and was continued monthly thereafter. MTX (maximum 10 mg/m2/week) and low-dose steroids (maximum 0.2 mg/kg/day), if present at baseline, were maintained or possibly tapered during the study period. Clinical data, collected at baseline and every month during followup, were the number of uveitis flares, new onset or worsening of preexisting ocular complications, and adverse events. A comparison between the frequency of uveitis flares during the 6-month period before the introduction of abatacept and the 6-month period after abatacept introduction was performed. The articular involvement was also recorded.

Laboratory monitoring included white blood cell count, hemoglobin level, platelet count, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level, alanine transaminase (ALT), aspartate transaminase (AST), creatinine level, blood urea nitrogen (BUN), urinalysis, antinuclear antibodies (ANAs), and anti–double-stranded DNA antibodies. ESR and CRP level were considered abnormal if they were >35 mm and >0.5 mg/dl, respectively; the white blood cell count and hemoglobin level were considered to be abnormal if they were greater or lower than 2 SD values for age. The platelet count was considered abnormal if it was >400 × 103/mm3. ALT and AST were considered abnormal if they were >40 units/liter, creatinine if it was >62μmoles/liter or >0.47 mg/dl, and BUN if it was >6.40 mmoles/liter. ANA titer, tested on the HEp-2 cell line, was considered to be positive if it was >1:80. These tests were conducted at the initiation of treatment and every 3 months during the followup

Informed consent was obtained by the patients if they were >18 years of age or by their parents or legal representatives if they were not. Because abatacept is one of the experimental treatment options with the potential to control and/or prevent further ocular complications, it was approved for compassionate use by the National Health Service Authority.


Seven patients, 6 females and 1 male, with oligoarticular JIA entered the study. All but 1 were ANA positive, and 2 were HLA–B27 positive.

The clinical and demographic characteristics of the patients are summarized in Table 1. The age at uveitis onset ranged from 3–14 years (mean 6.1 years), and the mean duration of chronic anterior uveitis at treatment initiation was 11.6 years (range 3–17 years). All patients had combinations of several ocular complications at baseline: band keratopathy (5 patients), posterior synechiae (4 patients), cataract (5 patients), vitreitis (3 patients), posterior vitreal detachment (1 patient), and cystoid macular edema (3 patients).

Table 1. Demographic and clinical characteristics of the patients*
 Patient no.
  • *

    JIA = juvenile idiopathic arthritis; CS = corticosteroids; MTX = methotrexate; CSA = cyclosporin A; MMF = mycophenolate mofetil; LEF = leflunomide; anti-TNF = anti–tumor necrosis factor; INFLIX = infliximab; ADA = adalimumab; ETAN = etanercept.

Age at JIA onset, years, months2, 08, 32, 64, 82, 61, 515, 0
Age at uveitis onset, years61435348
Uveitis duration, years177143151213
Ocular complicationsBand keratopathy, synechiae, cataract, posterior vitreal detachmentBand keratopathy, synechiae, cataractSynechiae, cataract, vitreitisBand keratopathy, vitreitis, cystoid macular edemaVitreitis, cystoid macular edemaBand keratopathy, synechiae, cataractBand keratopathy, cataract, cystoid macular edema
Previous steroid or immunosuppressive treatmentsCS, MTXCS, MTXCS, MTXCS, MTX, CSA, MMFCS, MTX, CSACS, MTX, CSA, MMFCS, MTX, CSA, LEF, MMF
Previous anti-TNF treatment, duration monthsINFLIX 34, ADA 25INFLIX 30, ADA 12ETAN 63, INFLIX 4, ADA 12ETAN 6, INFLIX 9INFLIX 30, ADA 27INFLIX 37, ADA 5ETAN 3, INFLIX 26, ADA 5
Withdrawal causeInefficacyInfusion reaction (INFLIX), inefficacy (ADA)InefficacyUveitis (ETAN), optic neuritis (INFLIX)InefficacyInefficacyInefficacy

In all cases, a step-up treatment approach for chronic anterior uveitis had been followed. Topical steroids and cycloplegic ophthalmic drops had been progressively administered with systemic steroids (oral or IV) and immunosuppressive agents (MTX as the first choice, followed, in 3 cases, by CSA or MMF). One patient was intolerant to MTX and was switched to LEF. Then, according to the disease severity, TNFα antagonists (infliximab or adalimumab) were used. Only 3 patients had been previously treated with etanercept for 3, 6, and 63 months for aggressive arthritis, and were then switched to infliximab due to the worsening of uveitis.

In 5 patients (patients 1, 3, 5, 6, and 7), the chronic anterior uveitis was refractory to both infliximab and adalimumab (Table 1). In 1 patient (patient 2), infliximab had been discontinued because of a severe infusion reaction and adalimumab was not effective. In 1 patient (patient 4), infliximab had been interrupted because of the onset of optic neuritis.

The uveitis outcome observed after abatacept treatment is shown in Table 2. The followup after abatacept initiation ranged 7–11 months (mean 9.2 months) in 6 patients. In 1 patient, abatacept was discontinued after 3 months of treatment because of skin reactions, oral mycosis, and arthritis flare. No infusion reactions or other drug-related adverse events were reported in the other patients.

Table 2. Results of the abatacept treatment*
 Patient no.
  • *

    RE = right eye; LE = left eye; lp = light perception.

Followup duration with abatacept, months9371081011
Abatacept-related side effectsNoneOral mycosis, skin reaction, arthritis flareNoneNoneNoneNoneNone
Uveitis flares before → after abatacept, no.3 → 04 → 14 → 03 → 14 → 05 → 23 → 0
Mean anterior chamber cells grading before → after abatacept2.6+ → 1.5+2.6+ → 2+2+ → 0.5+1.6+ → 0+4+ → 1.6+2.6+ → 1.3+2+ → 1+
Visual acuity 6 months before → 6 months after abataceptRE 0.63 → 1 LE 0.5 → 0.63RE 0.63 → 0.32 LE 0.63 → 0.8RE 0.4 → 0.2 LE 0.63 → 0.8RE 0.1 → 0.2 LE 0.2 → 0.1RE 0.05 → 0.1 LE lp → lpRE 0.32 → 0.4 LE 0.6 → 0.4RE lp → lp LE 0.8 → 1.0
Steroids tapering before → after abatacept, mg/day15 → 025 → 00 → 025 → 12.50 → 00 → 025 → 12.5

All patients had chronic active uveitis during the 6 months before the initiation of abatacept. All responded to abatacept treatment: in 3 patients, uveitis improved as early as 2 weeks after the abatacept start, in 3 patients after 1 month, and in 1 patient after 6 months. The frequency of uveitis flares decreased from a mean of 3.7 episodes (range 3–5 episodes) during the 6-month period before abatacept initiation to a mean of 0.7 episodes (range 0–2 episodes) during the 6-month period after abatacept initiation. Uveitis flares only occurred in 3 patients; they were mild and did not require an increase of steroid dose.

The mean anterior chamber cells grading significantly improved with abatacept, although only 1 patient went into complete remission. Visual acuity also improved but, unfortunately, the preexisting structural complications (band keratopathy, cataract, and cystoid macular edema) limited the complete recovery.

At the end of the followup, no new complications were observed and preexisting complications remained stable in all patients. Of note, among the 4 patients who were receiving corticosteroids at baseline, 2 were able to discontinue the treatment and 2 reduced their daily need by half (Table 2). All patients receiving MTX at baseline continued to receive it during the followup. In all but 1 patient (patient 4), the laboratory inflammatory indexes decreased or normalized during the followup. All of the other parameters (ALT, AST, creatinine, BUN, and urinalysis) remained within the normal range during the treatment.

During the washout period, 4 patients presented with persistent active arthritis and 3 experienced arthritis relapses. In all but 1 patient (patient 2), arthritis improved with abatacept, although more slowly than uveitis improved. During the followup, arthritis went into remission in 5 patients, and improved in 1 patient (patient 7) but persisted to be slightly active.


Several recent reports have shown that anti-TNFα agents represent a successful treatment for severe JIA-related uveitis in patients who do not respond to first-line immunosuppressive drugs. Etanercept was the first anti-TNFα agent used for chronic anterior uveitis, with controversial results. In fact, although a few reports have shown a significant association between etanercept and the development of new-onset uveitis and/or uveitis relapse (11–15), others have excluded this relationship (16).

Infliximab seems to be superior to etanercept in controlling chronic anterior uveitis (17), but serious side effects have been reported with infliximab treatment, including infusion reactions and optic neuritis. Adalimumab, a fully human anti-TNFα agent, seems to have comparable efficacy but better tolerability than infliximab; however, 20–35% of patients, usually those with a longer duration of JIA-related uveitis, seem to be resistant to this treatment (6).

The increasing knowledge of cell activation, particularly costimulatory signaling, has led to the development of new agents targeting cellular antigens, which play an important role in this process. Abatacept is a soluble fusion protein consisting of the extracellular domain of human CTLA-4 linked to the modified Fc domain of human IgG1. It avidly binds to CD80/CD86 on antigen-presenting cells and blocks the CD28 costimulatory signal, resulting in T cell inactivation (18). In particular, it has been shown that CTLA-4Ig inhibits the interaction between the CD80/CD86 and CD28 and, therefore, reduces the incidence and severity of autoimmune anterior uveitis in experimental models (19).

The safety and efficacy of abatacept in children with JIA who had failed previous treatment, including anti-TNF agents, were recently reported in a double-blind, randomized, controlled trial that excluded patients with active uveitis (8). Following from these results and an isolated case report that suggested abatacept as an effective choice for refractory uveitis (9), we evaluated whether abatacept might represent a good alternative treatment for severe JIA-associated uveitis that was refractory to immunosuppressive drugs and anti-TNFα agents.

All 7 patients in our series responded to abatacept treatment. The number of chronic anterior uveitis relapses was significantly reduced, and there were no new-onset ocular complications or worsening of preexisting ocular complications. The only patient who experienced treatment discontinuation had experienced previous cutaneous adverse reactions with anti-TNFα treatments. The response to abatacept was faster for uveitis than for arthritis, although, in the majority of patients (5 of 6), a mild inflammation in the anterior chamber was still present at the last followup, even if at a much lower degree than at baseline. Moreover, uveitis flares occurred in 3 patients receiving abatacept, but they were mild and did not require an increase of the steroid dose.

Despite the fact that this study included a small series of selected patients and had a relatively short period of followup, it clearly shows that abatacept may represent an effective and safe treatment for patients with severe anti-TNFα refractory uveitis. Indeed, this result, along with the biologic rationale showing that costimulatory blockade prevents T cell activation and TNFα production in experimental models of uveitis, opens new insights regarding further applications of abatacept in JIA-related uveitis. Future prospective studies are needed to confirm these preliminary findings.


All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Zulian had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Zulian, Balzarin, Martini, Cimaz, Zannin.

Acquisition of data. Zulian, Balzarin, Falcini, Martini, Alessio, Cimaz, Cimino, Zannin.

Analysis and interpretation of data. Zulian, Martini, Zannin.


We thank Fabio Vittadello, statistician, and Cristina Vallongo, MD (Department of Pediatrics, University of Padua, Padua, Italy), for their technical assistance and Carlo Salvarani, MD (Department of Internal Medicine, Rheumatology Unit, Reggio Emilia, Italy), for having referred one patient.