Infliximab Therapy for the Treatment of Refractory Ocular Inflammatory Disease FREE

Ocular inflammatory disease includes a wide spectrum of conditions such as chronic cicatrizing conjunctivitis, scleritis, and uveitis. Treatment of these sight-threatening problems has been greatly advanced with the application of traditional immunomodulatory therapy (IMT), including antimetabolites and alkylating agents.¹ However, a small yet significant proportion of patients remain refractory to conventional IMT. The recent development of biologic agents that target specific cytokines holds promise for additional effective treatments.²

Tumor necrosis factor α (TNF-α) is a cytokine and inflammatory mediator in animal models of uveitis.^3- 5 Its blockade is associated with decreased ocular inflammation in these animal models.^3- 5 In addition, TNF-α has been found in higher levels in the sera and aqueous humor of patients with uveitis.⁶ These findings form the rationale for the use of anti–TNF-α biologic agents in ocular inflammatory diseases.

Infliximab, a monoclonal antibody to TNF-α, is a biologic agent that has been approved for use in rheumatoid arthritis, Crohn disease, and ankylosing spondilitis.^7- 9 Many centers have used infliximab off-label for the treatment of different types of ocular inflammation with initially promising results.^10- 16 In this report, we describe our experience with the use of infliximab in patients with ocular inflammatory disease resistant to traditional IMT.

The Human Studies Committee of the Massachusetts Eye and Ear Infirmary approved the retrospective data review for this study with waiver of informed consent. Cases were consecutive patients with ocular inflammatory disease refractory to traditional IMT who received intravenous infliximab (Remicade; Centocor, Inc, Malvern, Pennsylvania) between January 1, 2001, and March 30, 2006, while under the care of 1 of us (C.S.F.) at a uveitis referral center. Patients at this center are treated with a stepladder approach to IMT, which typically starts with an antimetabolite, such as methotrexate, or a signal transduction inhibitor, such as cyclosporine. If patients fail this therapy, a switch to another medication in either of these 2 classes or a combination of antimetabolite and signal transduction inhibitor is tried. If this is insufficient, either an alkylating agent, such as cyclophosphamide, or a biologic agent is used. Therefore, in this study, refractory ocular inflammation generally refers to inflammation that was unresponsive to antimetabolite/signal transduction inhibitor combination therapy, alkylating drugs, or a biologic agent other than infliximab; treatment in these cases was given on an off-label–use basis for sight-threatening inflammation. However, in a few patients who had extraocular manifestations of disease, particularly arthritis, the patients' rheumatologists introduced infliximab therapy for the joint disease before therapy could be moved up the stepladder for ocular inflammation.

Infliximab infusions were administered to patients under the supervision of 1 of us (C.S.F.) or the patient's rheumatologist. Preinfusion blood tests included a complete blood cell count, liver function tests, serum creatinine level, and urea nitrogen level. A purified protein derivative skin test was performed in patients at risk for tuberculosis: health care workers and persons who had been in tuberculosis-endemic areas. Five milligrams per kilogram of infliximab were infused over 2 hours every 2 to 8 weeks. The infusion frequency was dependent on the indication for therapy and disease activity. If the patient was treated primarily for ocular inflammation, we treated every 4 weeks. If the infliximab therapy was initiated by the rheumatologist for concomitant inflammatory joint disease, the patient was treated by protocols established in rheumatologic clinical trials where infusions are given at weeks 0, 2, and 6 and then every 4 to 8 weeks.^7,9 In cases where joint disease but not eye disease was controlled with maintenance intervals greater than every 4 weeks, a joint decision with the rheumatologist was sometimes made to shorten the interval so that ocular inflammation could be eliminated as well. A complete blood cell count and liver function profile was obtained prior to each subsequent infusion. At each follow-up visit, a complete history and examination was performed.

Clinical information collected included previous and concomitant IMT, Snellen visual acuity, and degree of ocular inflammation. The information was recorded every 3 months for 1 year prior to and after initiation of infliximab therapy until the most recent visit, regardless of whether the patient discontinued infliximab use at some point. After the Standardization of Uveitis Nomenclature (SUN) Working Group grading scheme was established, it was used for anterior chamber cell and flare assesment.¹⁷ Otherwise, anterior chamber cell and flare, vitreous cell and haze, scleritis, and ocular cicatricial pemphigoid (OCP) activity were graded by previously described standardized grading criteria.^18- 20 The dose and frequency of infliximab infusions, adverse effects, and number of infusions were also recorded.

The primary outcome parameters were clinical response to treatment, measured as change in Snellen visual acuity and degree of inflammation, adverse events, and reduction in concomitant immunosuppression requirements. Snellen visual acuity change of 2 lines or more was considered a clinically significant change. In accordance with the SUN Working Group terminology, inactive anterior uveitis was defined as rare cells or less.¹⁷ For vitreous inflammation, scleritis, and OCP, inactive disease was defined as previously described.¹⁸ Unexpected complaints and complications were recorded as adverse events. The exceptions were mild flulike illness and mild injection site erythema. Concomitant IMT was tapered if the patient achieved control of inflammation. Any size decrease in the number or dosage of IMT medications was recorded.

Statistical analysis was performed using PROC LIFETEST in PC_SAS (version 6.08; SAS Institute, Cary, North Carolina). Because eyes were not examined independently and because disease progression and response to therapy are highly correlated between eyes,^{10,12,15,21- 22} we analyzed the data for left and right eyes separately.

The clinical data for each patient are summarized in Table 1. The ocular diagnoses included scleritis (n = 8), scleritis and anterior uveitis (n = 1), scleritis and panuveitis (n = 1), anterior uveitis (n = 8), panuveitis (n = 3), retinal vasculitis and panuveitis (n = 5), and OCP (n = 1). Twenty-one patients had underlying systemic diagnoses, which included rheumatoid arthritis (n = 2), juvenile idiopathic arthritis (n = 5), ankylosing spondylitis (n = 2), Behçet disease (n = 6), reactive arthritis (n = 2), relapsing polychondritis (n = 1), Crohn disease (n = 1), psoriasis (n = 1), and mucous membrane pemphigoid (n = 1). Five of these patients with systemic disease were HLA-B27 positive. The mean patient age was 37.3 years (range, 6-66 years). The mean number of infliximab infusions was 15.6 (range, 4-45), and the mean follow-up time from the first infusion was 25.6 months (range, 4-62 months). While all patients had active ocular inflammation, patients 1, 11, 14, 16, and 17 received infliximab primarily for joint inflammation and the remaining 22 patients received infliximab primarily for ocular inflammation.

The medications in Table 1 reflect the IMT regimen immediately prior to infliximab initiation. It does not include all of the IMT medications that each patient might have tried and failed previously. Patients 11, 14, 16, and 17 were only treated with methotrexate prior to the introduction of infliximab therapy; the infliximab was prescribed by their rheumatologists for joint disease and controlled their ocular inflammation before the usual stepladder approach to IMT could be continued for their eyes. Patient 23 was changed from methotrexate therapy directly to infliximab therapy because of early reports about Behçet disease responding well to anti–TNF-α therapy and because his serum TNF-α level was quite elevated at 27 pg/mL (reference range, 0-8.1 pg/mL). Tumor necrosis factor α levels were not checked in most patients and this is the only instance where the TNF-α level impacted therapy. Patients 2 and 18 were receiving monotherapy with methotrexate and prednisone, respectively, because of intolerance to multiple IMT medications.

The medications in Table 1 reflect the IMT regimen immediately prior to infliximab initiation. It does not include all of the IMT medications that each patient might have tried and failed previously. Patients 11, 14, 16, and 17 were only treated with methotrexate prior to the introduction of infliximab therapy; the infliximab was prescribed by their rheumatologists for joint disease and controlled their ocular inflammation before the usual stepladder approach to IMT could be continued for their eyes. Patient 23 was changed from methotrexate therapy directly to infliximab therapy because of early reports about Behçet disease responding well to anti–TNF-α therapy and because his serum TNF-α level was quite elevated at 27 pg/mL (reference range, 0-8.1 pg/mL). Tumor necrosis factor α levels were not checked in most patients and this is the only instance where the TNF-α level impacted therapy. Patients 2 and 18 were receiving monotherapy with methotrexate and prednisone, respectively, because of intolerance to multiple IMT medications.

Twenty-one of 27 patients achieved ocular inflammation control with the initial addition of infliximab therapy to their IMT regimens. Two additional patients (patients 3 and 22) achieved control of inflammation after an interval of alkylating agent therapy. Patient 3 had ongoing inflammation with her first trial of infliximab therapy, required intravenous cyclophosphamide therapy for several months, and then went back to infliximab and methotrexate therapy with good inflammation control since then. Patient 22 had recurrent inflammation while taking infliximab, requiring intravenous steroids followed by oral chlorambucil therapy for 4 months. After this “rescue” therapy, he has been inflammation-free while taking infliximab, cyclosporine, and prednisone for 8 months.

After achieving suppression of ocular inflammation, we were able to reduce either the dosage of a particular IMT medication or the number of IMT drugs for 9 patients (patients 2, 3, 5, 13, 15, 22, 23, 24, and 27). Seven patients (patients 4, 7, 8, 12, 16, 20, and 25) were able to discontinue all other IMT while receiving only infliximab therapy. Patient 9 received 4 infliximab infusions and then was noncompliant with visits. When he returned for care, his inflammation was controlled after not taking any medications for 7 months. The remaining 6 patients were unable to decrease either the dosage or number of IMT drugs.

Four patients did not achieve sustained inflammation control. Patient 17 responded so well to infliximab therapy initially that his methotrexate therapy was discontinued. Severe inflammation returned after 3 months, prompting reinstitution of methotrexate therapy. Control of his inflammation was eventually achieved with etanercept and methotrexate therapy without infliximab use. Patient 21 never achieved inflammation control while taking infliximab. Patients 11 and 18 achieved inflammation control initially but it eventually recurred while receiving infliximab.

Life-table methods were used to calculate cumulative incidences of inflammation control and vision change at 12 months (Table 2). Among eyes with inflammation at the first infusion (22 right eyes, 21 left eyes), greater than 90% showed complete resolution of inflammation at 12 months. Among eyes with a visual acuity of 20/30 or worse at the first infusion (10 right eyes, 13 left eyes), cumulative incidence of improvement in visual acuity by 2 or more Snellen lines at 12 months was 65% and 56% in right and left eyes, respectively. Cumulative incidence of a visual acuity decrease by 2 or more lines at 12 months (excluding 2 left eyes with light perception at baseline) was 12% for right eyes and 10% for left eyes.

Two patients noted headache and flushing initially with infliximab use, which resolved after continued treatments. Only patient 6 had an adverse event prompting infliximab discontinuation. She developed a lupuslike reaction with fatigue, photosensitivity, and a positive antinuclear antibody titer after 7 infusions. Infliximab use was discontinued, the symptoms resolved, and the antinuclear antibody titers normalized. The patient has not had a recurrence of the lupuslike syndrome or scleritis after 4 years of follow-up. The remaining 24 patients reported no adverse effects.

Patient 7 discontinued infliximab use after 7 infusions when she achieved control of inflammation. She has been inflammation-free for 17 months not receiving any IMT. Patient 8 discontinued infliximab use after 2 years of therapy and has been inflammation-free not taking any medications for 4 months. Patient 24 discontinued infliximab use after 6 months of remission, with ocular inflammation remaining quiet while taking low-dose prednisone for 3 years. Remission was achieved in patients 4 and 19 with infliximab use, but because of insurance noncoverage issues, infliximab use was discontinued. Patient 4's ocular inflammation remained quiet for 6 months after the last infliximab dose, without any additional IMT. Patient 19 achieved remission for 30 months after the last infliximab infusion while continuing therapy with mycophenolate mofetil. Patient 5 received 8 infliximab doses because of failure with azathioprine and prednisone use alone. His inflammation was controlled with infliximab use and he stopped the infusions because of the long travel distance from his rural home to the infusion center. He has remained inflammation-free for 5 months taking a lower dose of azathioprine and prednisone than what he was taking prior to infliximab initiation.

The rates of ocular inflammatory control with the initial addition of infliximab therapy in this study are 78% of patients and 81.4% of eyes, which are similar to the response rates reported in previous studies.^10,15 After additional courses of alkylating agents in 2 patients, the rates of inflammation control increased to 85% (23 of 27 patients) and 90.7% (39 of 43 eyes), respectively. This effect was sustained over a mean follow-up of 25.6 months, to our knowledge, the longest of any study published to date. This study includes patients who received and tolerated up to 45 infliximab infusions, to our knowledge, the largest number reported to date for ocular inflammatory disease. This suggests that infliximab therapy can be efficacious over the long-term. Thirteen patients continue to receive infliximab therapy on a regularly scheduled basis. The appropriate length of treatment for these patients is unclear and may be indefinite.

Life-table analysis further supported infliximab's efficacy for controlling ocular inflammation. These estimates do not take into account the ability of patients to comply with the therapy. Life-table analysis for a visual acuity improvement by 2 or more lines indicated that a majority of patients who had a visual acuity of 20/30 or worse at baseline had improved visual acuity after 1 year of treatment. The majority of our patients could not be included in this analysis because their visual acuity was 20/30 or better at the first infusion. This reflects our clinical approach, which is to generally treat patients with ocular inflammation before their vision is compromised. Life-table analysis for a visual acuity decrease indicated that less than 15% of patients had a drop in visual acuity by 2 or more lines at 1 year. This analysis included patients who failed infliximab therapy and had ongoing inflammation.

One patient (patient 7) with idiopathic scleritis had inflammation resolution after 7 infusions. She has remained relapse-free without IMT for 17 months. A second patient (patient 4) with rheumatoid arthritis–associated scleritis had to stop infliximab therapy after 6 infusions because of insurance noncoverage issues. She has remained inflammation-free for 6 months not taking any medications. A third patient (patient 8) stopped infliximab therapy after 2 years of inflammation control while receiving therapy, and he has been inflammation-free not taking any medications for 4 months. While biologic therapies are felt generally not to be curative, it is possible some patients may achieve at least short-term remission without the need for ongoing treatment. All 3 of these patients who achieved remission had scleritis as their ocular diagnosis.

Two patients who initially had inflammation recurrence while receiving infliximab therapy received a course of alkylating medication and were able to resume infliximab therapy with good inflammation control thereafter. Patient 3 received intravenous cyclophosphamide therapy for 6 months and patient 22 received 4 months of chlorambucil therapy. One relapse while taking infliximab should not eliminate it from the ocular immunologist's armamentarium for that patient. It may be useful as a maintenance drug after the acute episode of inflammation is quelled. In addition, given the recent finding that combination therapy with cyclophosphamide and etanercept, another anti–TNF-α biologic agent, may heighten the risk of cancer beyond that observed with cyclophosphamide therapy alone in the treatment of Wegener granulomatosis, patients should not be treated concomitantly with an alkylating agent and biologic agent until further information is available.²³ In this study of Wegener granulomatosis, however, prior therapy with cyclophosphamide did not affect the incidence of malignancy.

In contrast with the recent finding of a relatively high rate of adverse events in a prospective trial of infliximab therapy,¹⁰ our retrospective review included only 1 adverse event necessitating therapy discontinuation. As the authors of the aforementioned prospective trial point out, the causality for several of the significant adverse events is unclear.¹⁰ Other studies have shown low rates of adverse events similar to those in our report.^11,13- 14 Drug-induced lupus reaction is the one significant adverse event that we observed; this has been previously reported in the ophthalmic and rheumatologic literature with use of infliximab.^10,24 As with our patient, all the reported cases of lupuslike syndrome in clinical trials and postmarketing surveillance have improved with therapy discontinuation.²⁴

Our study included patients with a wide variety of ocular inflammatory conditions. Unlike other studies, we included patients with anterior uveitides of different etiologies as well as patients with posterior involvement. To our knowledge, this report includes the first published case of OCP treated with infliximab. Patient 27 reported subjective improvement of irritation and pain and had objective signs of improved inflammation after 8 infliximab infusions. Because of the control of his inflammation, he was able to undergo keratoprothesis surgery, which led to visual acuity improvement from light perception to 20/100 in the eye that underwent surgery. To our knowledge, our study also includes the largest series of patients with scleritis published to date. All 10 of these patients with scleritis had inflammation control while taking infliximab. Patients with scleritis may be a subgroup of patients who respond well to infliximab therapy and who are more likely to achieve remission. A larger randomized trial would be necessary to further test these hypotheses.

Four of our patients are continuing to receive infliximab therapy without any concomitant IMT. Typically, patients are maintained on low-dose immunosuppressive therapy to prevent antibody production to infliximab. The patients who opted for monotherapy were all informed of the risk of antibody development and subsequent tachyphylaxis to infliximab use. Their desire to avoid immunosuppressive therapy and their good inflammation control while taking infliximab led them to make this choice. All these patients have maintained an excellent response to infliximab therapy, having received between 20 and 45 infusions. Monotherapy may be offered to patients with uveitis and may be efficacious over the long-term.

In our experience, the optimal maintenance infliximab dosing interval for patients with uveitis is shorter than the 8 weeks that has been reported for patients with diseases such as Crohn disease.⁸ We have found that every-month dosing in these patients eliminates “escape” of inflammation between treatments, and this schedule is well tolerated. Of the 5 patients in this series who were treated primarily by their rheumatologists for joint disease with a prolonging of intervals between infusions up to every 8 weeks, 3 were able to achieve ocular inflammation control.⁷

We found infliximab therapy to be an efficacious biologic agent for the treatment of refractory ocular inflammation. There was a low incidence of adverse effects prompting discontinuation. This study, however, is limited by its retrospective nature, lack of a control group, relatively small number of patients, and limited follow-up period for some patients. Randomized controlled trials of infliximab therapy for patients with various subtypes of ocular inflammatory disease would be required to answer questions of efficacy and safety more definitively.

Correspondence: C. Stephen Foster, MD, Massachusetts Eye Research and Surgery Institute, 5 Cambridge Center, Eighth floor, Cambridge, MA 02142 (fosters@uveitis.org).

Submitted for Publication: June 10, 2006; final revision received November 20, 2006; accepted November 22, 2006.

Financial Disclosure: None reported.