Article Text
Abstract
Background/aims To evaluate retrospectively the long-term efficacy of rituximab in patients with severe juvenile idiopathic arthritis (JIA)-associated uveitis.
Methods Eight patients (15 eyes) with severe and longstanding JIA uveitis, who had an inadequate response in controlling uveitis to one or more biologic agents including tumour necrosis factor blockers and abatacept, received rituximab therapy. Rituximab was given at a dose of 1000 mg per infusion on days 1 and 15 and then every 6 months. Clinical responses to treatment, including decrease in uveitis activity, visual acuity changes, reduction of concomitant local and systemic corticosteroid and/or immunosuppressants, and occurrence of adverse events, were assessed.
Results Eight patients with a mean±SD age of 22.8±5.5 years were treated. The mean ocular disease duration was 17.7 years; the mean±SD follow-up time on rituximab was 44.75±4.9 months; and the mean number of rituximab infusions received was 8.75 (range 6–12). All patients achieved complete control of uveitis, but in two patients rituximab was discontinued due to inefficacy in treating arthritis. The decrease in uveitis activity was evident 4–5 months after the first infusion. Systemic corticosteroids and immunosuppressants used in association with rituximab were discontinued in five patients at the end of follow-up. None of the patients experienced visual worsening during the follow-up. No drug-related complications were encountered.
Conclusions Rituximab may be a promising effective treatment option for refractory uveitis associated with JIA leading to long-term quiescence of uveitis, particularly for patients who have not previously responded to other biologic therapies.
- Immunology
- Child health (paediatrics)
- Drugs
- Inflammation
Statistics from Altmetric.com
Introduction
Uveitis associated with juvenile idiopathic arthritis (JIA) is an insidious and severe inflammatory disease presenting as asymptomatic chronic anterior uveitis.1
Results from meta-analysis data suggest that the cumulative incidence of uveitis in JIA varies according to geographic location, being highest in Scandinavia, then the USA, then Asia, and lowest in India.2 The pooled data from this meta-analysis showed an incidence of 12.4% of pauciarticular JIA-associated uveitis and 4.3% in polyarticular patients.
Long-term visual prognosis in patients suffering from JIA-associated uveitis has been described as poor, and meta-analysis studies found a cumulative incidence of 9.2% of adverse visual outcome (visual acuity <20/40).2
Previous studies have estimated that the reported risks of reduced visual acuity to 20/200 range from 5% to 9% depending on series and practice type.3
Proper and aggressive immunomodulatory therapy is therefore critical to improve the long-term visual prognosis, and to reduce the risk of ocular complications and corticosteroid-induced adverse events.1 ,3 Since the introduction of biologic agents, tumour necrosis factor α (TNF-α) antagonists have notably improved the treatment options for JIA.4–7
We recently published our positive experience on the use of infliximab, adalimumab,8 and golimumab9 in Italian patients with JIA associated uveitis.
However, a subset of patients fails to respond to TNF-α blockers or is intolerant to these therapies and may benefit from switching to a different biologic agent.10 Rituximab, the anti-CD20 B cell monoclonal antibody therapy, has been used successfully in the treatment of rheumatoid arthritis (RA) and other autoimmune diseases that have responded inadequately to TNF-α blockers.11 ,12
Few data are present in the literature on the efficacy and tolerability of rituximab treatment in patients with ocular inflammatory diseases13 ,14 and, particularly, very little data are available for patients with JIA uveitis.15–18
We previously reported our favourable preliminary experience in treating patients with JIA-associated uveitis with rituximab,15–18 and the cases presented herein have a longer follow-up time of observation.
The purpose of the present study was to evaluate retrospectively the long-term efficacy of rituximab in eight patients with severe and recalcitrant JIA-associated uveitis who failed to respond to other biologic therapies.
Methods
This is a retrospective study of the case records of eight patients with JIA-associated uveitis who received rituximab therapy from June 2009 to December 2014.
The Ethics Committee ruled that approval was not required for this study and the study was conducted in adherence with the declaration of Helsinki after obtaining written informed consent from each patient.
Patients were followed at the Paediatric Rheumatologic Unit at the ‘Istituto Ortopedico G. Pini’, Milan and were always evaluated by the same uveitis specialist (EM). Inclusion criteria for treatment with rituximab were inadequate response in controlling uveitis with traditional immunosuppressives and at least one TNF-α inhibitor (etanercept, infliximab or adalimumab), and/or abatacept.
Rituximab was given at a dose of 1000 mg per infusion on days 1 and 15 and a recall infusion was scheduled approximately after 6 months upon activity of uveitis and/or arthritis, according to the rheumatologic protocol applied in the treatment of RA. All patients received 100 mg of methylprednisolone intravenously, oral paracetamol and antihistamines before rituximab infusion, to reduce the risk of infusion reactions. A fixed 6-month re-treatment protocol was used in all patients.
A complete serological evaluation to rule out the presence of active or latent tuberculosis was performed in all patients. Serum biochemical and haematologic profiles with complete haemogram, liver and renal function tests, peripheral blood B lymphocyte count, chest X-ray, and electrocardiography were carried out before initiation of rituximab treatment.
CD19/20 B cell counts were performed at baseline before initiation of rituximab treatment and then were not regularly repeated in each patient on a fixed protocol. B cell depletion was confirmed by flow cytometry (less than one CD19+ cell/µl). However the decision to repeat rituximab infusions was based only on the clinical response of uveitis and arthritis and not on the level of CD19 cells.
Ophthalmologic evaluation and rheumatologic assessments were performed every 1 or 2 months according to disease activity and response to treatment. Complete physical examination and routine laboratory tests were repeated at each visit by the same paediatric rheumatologist (IP).
Clinical parameters evaluated were: age, gender, age at onset of uveitis and arthritis, characteristics of the uveitis, ocular complications, JIA category according to International League of Associations for Rheumatology (ILAR) classification,19 presence of antinuclear antibodies (ANA), rheumatoid factor (RF) and HLA-B27 antigen, previous systemic immunosuppressive therapies, previous local corticosteroid therapy, number and dosage of rituximab infusions, and follow-up on rituximab treatment.
The outcome measures analysed in the study were: clinical response to treatment including improvement in uveitis activity (two step decrease in level of anterior chamber or vitreous inflammation or decrease to grade 0 cells), visual acuity improvement, tapering of concomitant local and systemic medications (corticosteroids and/or immunosuppressant), and occurrence of adverse events. Disease activity was graded in accordance with the standardisation uveitis nomenclature (SUN) criteria.20 Anterior chamber cells, as a validated indicator for inflammatory activity, and Snellen visual acuity were graded before treatment and at every ophthalmologic assessment.
Results
Eight Caucasian patients (two males and six females; 15 affected eyes) with a mean±SD age of 22.8±5.5 years (range 16–34 years) affected by JIA and associated uveitis were treated with rituximab.
Indication for treatment was active uveitis and failure of conventional immunosuppressants and biologic therapy including TNF-α blockers (etanercept, infliximab, adalimumab) or abatacept due to inadequate response or occurrence of adverse events. All patients received rituximab treatment for active uveitis and/or active arthritis. Ocular involvement and demographic data are summarised in table 1.
Five patients were affected by oligo-extended JIA and three by oligo-persistent JIA according to ILAR classification. All patients were ANA positive, and negative for RF and HLA-B27 antigen. The mean±SD age at onset of arthritis was 3.3±2.2 years (range 1–8 years), while the mean age at onset of uveitis was 4.7±3.6 years (range 1–12 years); the mean ocular disease duration was 17.7±5.6 years (range 11–26 years). Seven patients had bilateral ocular involvement (total number of treated eyes: 15). Anterior uveitis was the most common type of uveitis and was present in five patients, while three patients presented also with posterior pole complications. Ocular complications present at baseline, before initiation of rituximab treatment, were: posterior synechia (13 eyes), cataract (5 eyes), band keratopathy (13 eyes), glaucoma (6 eyes), epiretinal membrane (2 eyes), cystoid macular edema (1 eye), and phthisis bulbi (2 eyes). No additional structural complications were observed during follow-up on rituximab. Table 2 summarises the details of previous and concomitant medical treatment for each patient.
Before starting rituximab treatment, all patients had been unsuccessfully treated with different consecutive TNF-α blockers or abatacept. Patients 5 and 8 also received a short-term course of chlorambucil before starting the biologic therapy.
The mean dose of systemic prednisolone needed before and after rituximab treatment decreased from 17.18 mg/day (range 7.5–25 mg) to 1.8 mg/day at last evaluation (range 0–7.5 mg). At the end of follow-up, only three patients were still on daily systemic low doses of prednisolone (2.5–7.5 mg).
The dosage of immunosuppressants used in association with rituximab was reduced during the follow-up: methotrexate was discontinued in four patients and ciclosporin in three patients at the end of follow-up (table 2). The number of daily topical corticosteroid drops was reduced from a mean of 3.6 (range 2–6) times a day to a mean of 0.37 (range 0–1) times a day after rituximab therapy, and five patients were able to discontinue topical therapy at their last visit. The clinical response of uveitis to rituximab is presented in table 3.
At last follow-up visit all eight patients achieved complete control of the uveitis and presented with inactive uveitis.
However, two patients (4 and 5) discontinued rituximab after 29 and 26 months because of its inefficacy in treating arthritis; despite uveitis responding well in both patients, the arthritis did not respond to rituximab and the patients were switched to golimumab.
The mean number of rituximab infusions that patients received during follow-up was 8.75 (range 6–12). No serious adverse events or rituximab-related complications were encountered in our treated patients during the long-term follow-up.
Recurrences of anterior uveitis experienced by patients during follow-up are listed in table 3. Patients with relapse of uveitis while on rituximab were treated with short-term topical corticosteroids according to the degree of intraocular inflammation. The mean number of uveitis recurrences decreased from 0.7 episodes per year before rituximab treatment to 0.2 episodes per year at last examination.
Improvement in uveitis (two step decrease in level of anterior chamber cells or vitreous haze) was evident 4 months after the first rituximab infusion.
The reduction of cystoid macular oedema in patient 2 was noticeable 3 months after rituximab infusion, with evidence of reduction of central macular thickness on the optical coherence tomography exam.
The mean±SD uveitis activity before treatment was 2.7±0.4 cells and 0.4±0.3 cells at last follow-up, while the flare values before and after rituximab were 3.5±0.4 and 0.8±0.3, respectively.
The mean follow-up time on rituximab was 44.75±4.9 months (range 26–62 months). After exclusion of the two patients who were non-responders for arthritis, the mean follow-up time on rituximab of the remaining six patients was 50.5 months (range 27–62 months).
Snellen visual acuity measurements of each patient at the beginning of rituximab treatment and at last visit are shown in table 3. None of the patients had a visual worsening during the follow-up, while one (patient 2) had a visual acuity improvement in the right eye following cataract surgery while on clinical remission on rituximab treatment. No other surgical interventions were performed during treatment. The lack of visual improvement (eyes with no light perception) was due to severe posterior pole complications at baseline such as phthisis bulbi, glaucomatous optic neuropathy and epiretinal membrane.
Discussion
In the present study we report our clinical experience on the long-term treatment with rituximab in eight patients with severe and recalcitrant JIA-associated uveitis. Rituximab is a chimeric murine/human monoclonal antibody against CD20 antigen that is expressed on the B cell surface.11 ,12 ,21
Little is known about the use of rituximab in ophthalmology,13 ,14 particularly in JIA-associated uveitis.15–18 Based on the favourable response in inducing long-term remission in different ocular and rheumatic inflammatory diseases, we previously published our positive preliminary results with rituximab in the treatment of patients with JIA uveitis.16–18
In the present long-term study we report the favourable efficacy of rituximab in eight patients treated for a mean follow-up time of 44.75 months. At last follow-up visit the uveitis was controlled in all patients, but two patients necessitated discontinuation of rituximab because of uncontrolled active arthritis. After exclusion of these two non-responders, the mean follow-up time on rituximab of the remaining six patients was 50.5 months (range 27–62 months).
To date, this is the first long-term experience on the treatment of JIA-associated uveitis with anti-CD20 monoclonal antibody.
The cases included in the present series were the most recalcitrant forms of JIA uveitis followed in our tertiary care; this is confirmed by the longstanding duration of uveitis (mean duration 17.7 years), and by the high number of ocular complications at the beginning of treatment. Patients were recalcitrant to conventional immunosuppressants or different types of biologic agents, as demonstrated by the multiple ‘switch’ of biologics employed. TNF-α blockers and abatacept were discontinued due to lack of efficacy in treating uveitis and/or occurrence of unacceptable treatment-related side effects. Although long-term remission with chlorambucil had been previously reported with JIA-associated uveitis,22 short-term use of the agent failed to control uveitis in two of our patients.
Another supportive element that confirms the severity of the uveitis in our group is the relative older age (mean age 22.8 years; range 17–34 years), indicating that these patients represent a subset of patients with longstanding uveitis requiring aggressive treatment starting from childhood and continuing to adult life.
Similar to other reports, in the rheumatologic experience with rituximab, improvement of intra-ocular inflammation with reduction of anterior chamber cell reaction (positive clinical response) was noticeable around the fourth month after the first infusion. Nevertheless, how the dose and dosing interval may be adapted during long-term treatment has not yet been established.11 ,12
During our long-term follow-up a considerable corticosteroid sparing effect was observed: in six patients systemic corticosteroids were completely discontinued, while in two other patients only low daily doses (2.5 to 5 mg) were still required to control the disease.
Associated conventional immunosuppressants were also tapered during treatment with rituximab upon clinical remission of uveitis and arthritis, and only two patients among the responders maintained methotrexate weekly therapy until the last follow-up visit.
Rituximab was well tolerated in all patients and we did not observe any side effects or drug-related complications during the mean follow-up time of 44.75 months. The tolerability and safety of rituximab has been well described in clinical trials of patients with RA and non-Hodgkin's lymphoma.12 ,18 ,23
The main concern with the use of rituximab in the treatment of autoimmune diseases, especially when used with concomitant immunosuppressive therapy or high dose corticosteroids, are infections.23 However, the risk of infectious complications has been reported as being low in patients in whom rituximab has been administered alone without systemic corticosteroids,24 such as in our JIA patients. Although in our experience rituximab therapy has had a favourable adverse reaction profile in JIA patients, we should take into consideration the potential serious side effects reported in adult patients with autoimmune systemic diseases and limit rituximab therapy for treatment-resistant patients with ocular inflammation. The therapeutic approach in our JIA patients was different from adults,12 repeating infusions every 6 months to maintain long-term quiescence and avoid ocular recurrences of uveitis.
Results from our clinical experience are encouraging and indicate that rituximab treatment can result in long-term disease quiescence, allowing discontinuation of other immunosuppressive and corticosteroid treatment. However, several factors must be taken into consideration when treating patients with JIA-associated uveitis: firstly, there is an important variability in the response to the biologics among these patients; secondly, the decision to treat with an alternative biologic agent is based on retrospective analyses from case series predominantly based on rheumatological experience; and thirdly, even when administration of these agents appears to be beneficial, in several patients the efficacy of the biologics in treating uveitis and arthritis may be different, necessitating a switch to another therapeutic agent.
In conclusion, rituximab may be a new therapeutic option with a convenient dosing schedule for the treatment of patients with autoimmune diseases, particularly those who have not previously responded to TNF-α blockers. Our results must be interpreted with caution in the context of the retrospective design, the small number and heterogeneity of the patients, and the lack of a control group. Further randomised, prospective trials of rituximab in larger numbers of patients are needed to evaluate better the efficacy, dosing regimen, and safety of this treatment in patients with JIA-associated uveitis recalcitrant to previous conventional treatments.
References
Footnotes
Contributors Design of the study (EM, GM, IP, VG); analysis and interpretation (EM, GM, LB, IP, VG); writing the article (EM, LB, VG); critical revision of the article (EM, GM, LB, IP, PM, VG); final approval of the article (EM, GM, PM, VG); data collection (EM, LB, IP, VG); provision of materials, patients, or resources (EM, GM, LB, PM, VG); statistical expertise (EM, LB); literature search (EM, GM, VG); administrative, technical, or logistic support (EM, LB).
Competing interests None declared.
Patient consent Obtained.
Ethics approval Ethics Committee ruled that approval was not required for this study
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- At a glance