Long-Term Outcomes of Treatment with Biological Agents in Eyes with Refractory, Active, Noninfectious Intermediate Uveitis, Posterior Uveitis, or Panuveitis

Long-Term Outcomes of Treatment with Biological Agents in Eyes with Refractory, Active, Noninfectious Intermediate Uveitis, Posterior Uveitis, or Panuveitis

Long-Term Outcomes of Treatment with Biological Agents in Eyes with Refractory, Active, Noninfectious Intermediate Uveitis, Posterior Uveitis, or Panu...

506KB Sizes 0 Downloads 61 Views

Long-Term Outcomes of Treatment with Biological Agents in Eyes with Refractory, Active, Noninfectious Intermediate Uveitis, Posterior Uveitis, or Panuveitis Ahmed Al-Janabi, MSc, PhD,1 Amgad El Nokrashy, MSc, MD,1,2 Lazha Sharief, MPH, PhD,1 Vivekka Nagendran, BA, MBBS,1 Sue Lightman, FRCOphth, PhD,1 Oren Tomkins-Netzer, MD, PhD1,3 Purpose: To examine a large cohort of patients treated with biologic agents for active noninfectious intermediate uveitis, posterior uveitis, or panuveitis (NIPPU) and to compare their efficacy and long-term effect. Design: Retrospective, longitudinal study. Participants: Eighty-two patients (156 eyes) with active NIPPU after failure of treatment with corticosteroids and a second-line immunosuppression drug and treated with biologic agents who were treated at Moorfields Eye Hospital between 2001 and 2016. Methods: Information was gathered from the clinical notes of all patients. Main Outcome Measures: Time to first disease flare, rate of treatment failure, best-corrected visual acuity, and risk factors for treatment failure. Results: Patients were followed on average for 4.70.4 years (724 eye-years). All patients demonstrated active uveitis at baseline, and 34 patients (41.5%) demonstrated a coexisting active systemic disease. Control of ocular inflammation was achieved in 136 eyes (87.2%). The average oral prednisolone dose at baseline was 16.41.7 mg/day, and by 6 months reduced to 6.50.7 mg/day (P < 0.0001), remaining stable for up to 5 years follow-up. Best-corrected visual acuity at baseline was 0.50.1 logarithm of the minimum angle of resolution (logMAR), improved to 0.40.1 logMAR (P ¼ 0.008) at 3 months, and remained stable during follow-up. After baseline, 42.3% of eyes experienced flares, and the average number of flares reduced from 1.80.1 flares/year to 0.60.1 flares/year (P < 0.0001). Median time to first flare was 5.4 years (95% confidence interval [CI], 2.2e5.4 years) with a 5-year survival rate of 58.7%. Treatment failed in 37 eyes (23.7%), with a 5-year survival rate of 68.0% and an estimated time to 75% survival of 2.9 years (95% CI, 2.1e4.4 years). The risk for treatment failure was lower when treatment used adalimumab (odds ratio, 0.4; 95% CI, 0.2e0.9; P ¼ 0.03) but was greater when systemic disease also was active at baseline (odds ratio, 3.2; 95% CI, 1.5e7.1; P ¼ 0.004). Conclusions: Overall, eyes treated with biologic agents after failure of treatment with corticosteroids and a second-line immunosuppression drug experienced satisfactory disease control (87.2%), reduced use of systemic immunosuppression, stable visual acuity, and a 23.7% risk of disease relapse. After multivariate adjustment, older age, treatment with adalimumab (versus infliximab), and inactive concomitant systemic disease were associated with a lower risk of treatment failure. Ophthalmology 2019;-:1e7 ª 2019 by the American Academy of Ophthalmology

The noninfectious uveitides are a group of ocular immunemediated inflammatory diseases, either part of systemic conditions such as sarcoidosis, Behçet’s disease, or juvenilerelated arthritis or isolated ocular immune-mediated disease.1 Treatment of these conditions requires the long-term use of immunosuppression agents, primarily corticosteroids, aimed at controlling the inflammation and maintaining vision.2 Treatment of noninfectious intermediate uveitis, posterior uveitis, or panuveitis (NIPPU) is undergoing significant advances with the development of biologic drugs directed at specific targets in the inflammatory cascade.3 The introduction of antietumor necrosis factor a agents and subsequent licensing of adalimumab for treating ª 2019 by the American Academy of Ophthalmology Published by Elsevier Inc.

NIPPU have expanded the range of drugs at our disposal and offer patients with refractory disease additional treatment options. Although the use of biologic drugs is increasing for treating patients with refractory, active disease, questions remain regarding their effect as long-term treatment, in particular, in maintaining vision and allowing reduction of corticosteroids and other immunosuppressive drugs. The effect of biologic drugs on NIPPU may be variable, with evidence suggesting that some (such as rituximab) have little effect on intraocular inflammation or (like etanercept) may even increase the risk of disease relapse.4,5 Even with regard to the use of infliximab and adalimumab, which are https://doi.org/10.1016/j.ophtha.2019.08.031 ISSN 0161-6420/19

1

Ophthalmology Volume -, Number -, Month 2019 considered to be effective for treating NIPPU, it remains unclear if there is a therapeutic advantage in choosing one or the other.6e9 The recent evidence originating from the VISUAL studies provides significant support for the role of biologic agents in treating noninfectious uveitis. The results of the trials demonstrated that adalimumab offers an advantage in disease control for patients with NIPPU, reducing the risk of relapse and allowing for the reduction in other immunosuppression drugs, primarily corticosteroids.10e12 However, the relatively short length of follow-up provided only partial data regarding the risk of disease developing and time to disease flare and treatment failure. In this study, we examined eyes of patients with either active or quiescent systemic disease and refractory, active NIPPU that previously had failed treatment with corticosteroids and second-line agents. We examined the effect of treatment with biologic drugs on sustained disease control, visual acuity, and risk of treatment failure and compared these between different agents.

Methods This was a retrospective, longitudinal study of patients with refractory NIPPU diagnosed between 2001 and 2016 and treated with biologic agents at Moorfields Eye Hospital, London, United Kingdom. The study adhered to the tenets of the Declaration of Helsinki and was conducted after ethical approval for data collection (identifier, ROAD16039, Moorfields Eye Hospital). Because this was a retrospective study, informed consent was not required. Patients were identified from a cohort of 1258 eyes with NIIPU. The study included patients of all ages who were diagnosed with NIPPU, whose disease failed treatment with corticosteroids and at least 1 second-line immunosuppression drug, demonstrated active uveitis at the time treatment with a biologic agent was started, and had at least 3 months follow-up after beginning the biologic agent. Patients were excluded if they showed anterior uveitis, uveitis resulting from an infectious cause, a known demyelinating disease before beginning biologic treatment, or if follow-up was less than 3 months. Adalimumab treatment was always started with a loading dose of 80-mg subcutaneous injection given at baseline followed by 40 mg every other week. The dose of intravenous infliximab was calculated based on patient weight at 5 mg/kg and given at baseline and weeks 2 and 6 and every 6 weeks after that. Before beginning biologic treatment, patients with intermediate uveitis underwent a review of systems, including any signs and symptoms related to multiple sclerosis. Further investigations and treatment were administered as needed.

Patient Selection Patient demographics and clinical information were collected for the entire follow-up period and were entered into a database. This included length of follow-up, anatomic diagnosis of uveitis using Standardization of Uveitis Nomenclature (SUN) criteria,13 treatment, response to treatment, time to and number of flares, ocular complications, and best-corrected visual acuity (BCVA) throughout follow-up. Baseline was defined as the time that treatment with the first biologic agent was initiated. Disease control was defined if the time to first flare was more than 3 months from baseline. Disease flare was defined as the diagnosis of intraocular inflammation occurring in the same eye at least 6 months after inflammation was controlled and requiring the reintroduction or

2

addition of another immunosuppression drug. Treatment failure was defined when repeat flares (2) or systemic side effects resulted in stopping or switching a biologic drug. Best-corrected visual acuity was recorded using Snellen charts and was converted to logarithm of the minimum angle of resolution for statistical analysis. Best-corrected visual acuity, oral prednisolone dose, and second-line immunosuppression drug use were recorded at predefined time points: baseline and 3, 6, 9,12, 24, 36, 48, 60, 72, 84, 96, 108, and 120 months. Stable oral prednisolone dose was defined as a dose not statistically different from that at 6 months of follow-up. A therapeutic dose for second-line immunosuppression drugs was defined as 1000 mg/day or more for mycophenolate mofetil, 10 mg/week or more for methotrexate, 150 mg/day or more for azathioprine, 3 mg/kg daily or more for cyclosporine, and 3 mg/day or more for tacrolimus. Moderate vision loss was defined as BCVA less than 20/40 and more than 20/200, and severe vision loss was defined as BCVA of 20/200 or less.

Statistical Analysis Comparing the change in mean BCVA and number of flares per year throughout follow-up was performed using generalized estimating equations, adjusting for time of follow-up, correlations between 2 treated eyes of the same patient, and incomplete data. Change in average dose of prednisolone was performed using a 1-way analysis of variance and a Bonferroni post hoc analysis, accounting for incomplete data. Percent of patients achieving a prednisolone dose of less than 10 mg/day and a nontherapeutic dose of second-line immunosuppression drugs was carried out using the Fisher exact test. Multivariate analysis and odds ratios with 95% confidence intervals (CIs) for treatment failure were calculated using a Cox regression model while adjusting for correlations between both eyes of the same patient and accounting for the variable follow-up. Only factors with a significance level of at least 0.1 on univariate analysis were included in the multivariate model. The Kaplan-Meier estimator was used to examine survival from the first flare, treatment failure, and achieving a prednisolone dose of less than 10 mg/day. As soon as an event occurred, further information was censored. Groups were compared using a log-rank test. SPSS software version 24 (IBM, Chicago, IL) was used for all analyses. The accepted level of significance for all tests was a  0.05. Continuous data are presented as mean  standard error of the mean (SEM).

Results From the cohort of 1258 eyes with NIPPU, 156 eyes (80 right) of 82 patients (32 female) matched the inclusion criteria for this study. The average age at the time of diagnosis was 33.91.7 years (range, 4.4e69.5 years). There were an additional 4 patients whose treatment was stopped after less than 3 months because of side effects and therefore were not included in the study. Average age at the time biologics were started (baseline) was 37.81.8 years (range, 5.5e69.9 years). Length of follow-up after baseline was 4.70.4 years (range, 0.3e16.6 years; 724 eye-years). For 29 patients (53 eyes [34.0%]), follow-up was longer than 5 years. Eyes were diagnosed with intermediate uveitis (n ¼ 42), posterior uveitis (n ¼ 31), or panuveitis (n ¼ 83), including those diagnosed with HLA-B27erelated uveitis, 13 of whom demonstrated panuveitis and 11 of whom demonstrated intermediate uveitis. The first biologic agent used was infliximab for 76 eyes, adalimumab for 60 eyes, etanercept for 4 eyes, rituximab for 14 eyes, and vedolizumab for 2 eyes. At baseline, all eyes demonstrated active uveitis, although in 34 patients (41.5%), biologic treatment also was started because of a coexisting active systemic disease. Control

Al-Janabi et al



Biologics and Noninfectious Uveitis

Table 1. Patient Demographics Demographic Female gender, no. (%) Right eye, no. (%) Age at baseline (yrs), mean  SEM Prednisolone dose at base line (mg/day), mean  SEM Anatomic type of uveitis, no. (%) Intermediate uveitis Posterior uveitis Panuveitis Cause, no. of eyes (%) Behçet disease Vasculitis HLA-B27erelated uveitis Vogt-Koyanagi-Harada syndrome Sarcoidosis Systemic lupus erythematosus Serpiginous choroiditis Takayasu vasculitis Punctate inner choroidopathy Rheumatoid arthritis Multifocal choroiditis Idiopathic Juvenile idiopathic arthritis Blau syndrome Birdshot chorioretinopathy First biologic drug, no. of eyes (%) Infliximab Adalimumab Time to prednisolone <10 mg/day (mos), median (95% CI) Infliximab Adalimumab

Data 32 (39.02) 80 (51.3) 37.81.8 16.41.7

42 (26.9) 31 (19.9) 83 (53.2) 51 8 25 6 8 3 2 1 2 4 2 34 6 2 2

(32.7) (5.1) (16.0) (3.8) (5.1) (1.9) (1.3) (0.6) (1.3) (2.6) (1.3) (210.8) (3.8) (1.3) (1.3)

76 (48.7) 60 (38.5)

6.0 (3.37-8.63) 3.0 (1.77-4.23)

CI ¼ confidence interval; SEM ¼ standard error of mean.

of the ocular inflammation was achieved in 136 eyes (87.2%). Table 1 summarizes the demographic information regarding eyes treated with biologic agents. Most eyes were treated using infliximab (n ¼ 76 [48.7%]) and adalimumab (n ¼ 60 [38.5%]), and our analysis focused on these eyes. Between 2011 and 2013, eyes were more likely to be prescribed infliximab than adalimumab (n ¼ 59 [69.4%] vs. n ¼ 26 [30.6%], respectively); from 2014 through 2016, eyes were more likely to be prescribed adalimumab than infliximab (n ¼ 34 [66.7%] vs. n ¼ 17 [33.3%], respectively). At baseline, 56 patients (71.8%) were receiving oral prednisolone at a dose of more than 7.5 mg/day. By 6 months from baseline, 46 patients (59.7%; P < 0.0001; Fig 1A) managed either to reduce the prednisolone dose to 7.5 mg/day or less (n ¼ 25 [32.5%]) or to stop it completely (n ¼ 21 [27.3%]). This continued to increase, and by 5 years, 76% of patients (P < 0.0001) were receiving a prednisolone dose of 7.5 mg/day or less (n ¼ 8 [32%]) or had stopped it completely (n ¼ 11 [44%]). The average prednisolone dose at baseline was 16.51.28 mg/day, reduced by 6 months to 6.40.5mg/day, and it continued to remain stable for up to 5 years of follow-up (P < 0.0001; Fig 1B). Among those with follow-up longer than 5 years, prednisolone dose continued to remain stable and was 6.72.7 mg/day at 10 years. For the entire cohort, median time to achieving a prednisolone dose lower than 10 mg/day was 3 months (95% CI, 1.2e4.8; Table 1). Patients receiving adalimumab achieved this at a median of 3 months, a significantly shorter interval than those treated with

Figure 1. A, Bar graph showing percent of patients receiving oral prednisolone (Pred) dose 10 mg/day or more (black bars) or less than 10 mg/day (white bars). After treatment with a biologic agent, significantly more patients were able to reduce the prednisolone dose to less than 10 mg/day or stop it completely. B, Graph showing that after treatment with a biologic agent, the average prednisolone dose reduced significantly by 3 months and remained stable at 5 years. C, Graph showing that visual improved after baseline and remained stable throughout follow-up. *P < 0.05. **P < 0.01. BCVA ¼ best-corrected visual acuity; logMAR ¼ logarithm of the minimum angle of resolution.

3

Ophthalmology Volume -, Number -, Month 2019 Table 2. Causes of Vision Loss at Final Follow-up Moderate Vision Loss, No. (%) Macular scarring ERM Amblyopia RD CME Macular ischemia Optic neuropathy Phthisis Total

9 1 1 2 6 1 1

(6.6) (0.7) (0.7) (1.4) (4.4) (0.7) (0.7) 0 22 (16.18)

Severe Vision Loss, No. (%) 10 0 0 2 5 1 1 1 20

(7.4) (0) (0) (1.4) (3.7) (0.7) (0.7) (0.7) (14.71)

CME ¼ cystoid macular edema; ERM ¼ epiretinal membrane; RD ¼ retinal detachment.

infliximab (6 months; P ¼ 0.04). Before baseline, 75 patients (91.5%) received treatment with a second-line immunosuppression drug, 44 were treated using 1 drug, 30 were treated with 2 drugs, and 1 patient received 3 drugs. At baseline, 50 of these patients (71.4%) were still receiving second-line immunosuppression drugs, 22 were treated with mycophenolate mofetil, 18 were treated with methotrexate, 6 were treated with azathioprine, 3 were treated with cyclosporine, and 1 was treated with tacrolimus. Of those receiving a second-line immunosuppression drug, 47 patients (94%) were receiving treatment at a therapeutic dose. By 12 months of follow-up, 44.7% of patients either had stopped treatment with a second-line agent or had reduced treatment to a nontherapeutic dose (P <0.0001). The percentage of patients managing to stop treatment continued increasing to 48.5% and 62% by 12 and 24 months follow-up, respectively (P <0.0001). Best-corrected visual acuity at baseline was 0.50.1 logMAR, improving significantly to 0.40.1 logMAR at 3 months (P ¼ 0.008) and remaining unchanged for the remainder of followup (Fig 1C). At 5 years, there seemed to be a deterioration in average visual acuity (0.60.1 logMAR), but the number of eyes (n ¼ 39) was too small to determine statistical significance (P ¼ 0.4). Although statistically significant, the improvement is not clinically significant; therefore, BCVA was regarded as stable. Among those with follow-up longer than 5 years, BCVA continued to remain stable at 0.30.1 logMAR by 10 years. At baseline, 81 eyes (62.3%) showed BCVA better than 20/40, 28 eyes (21.5%) already demonstrated moderate vision loss, and 21 eyes (16.2%) demonstrated severe vision loss (BCVA, 20/200). Vision loss rates improved by 24 months, with 77.4% showing a BCVA of better than 20/40 (P ¼ 0.02). There was no difference in the rates of vision loss at all other time points. At final follow-up, macular scarring was the main cause of vision loss among 22 and 20 eyes with moderate vision loss or severe vision loss, respectively (Table 2). To examine the effectiveness of biologic agents in refractory, active NIPPU, we analyzed the rates of disease flares after treatment. Before baseline, eyes experienced an average of 4.80.4 flares (2.00.2 flares/year), whereas after baseline, this reduced to an average of 1.70.2 flares (0.60.1 flares/year; P < 0.0001). The estimated median time to first flare was 5.4 years (95% CI, 2.2e5.4; Fig 2A), with a survival rate of 58.7% by 5 years and a cumulative 41.9% of eyes (n ¼ 57) flaring. Of the eyes treated with adalimumab, 29 (48.3%) experienced a flare at an average time of 1.20.2 years, compared with 28 eyes (36.8%) treated with infliximab that experienced a first flare at 1.00.2 years. Treatment with biologics failed in 24.3% of eyes (n ¼ 33) during follow-up (Fig 2B) with a 5-year survival rate of 68.0% and an estimated time to 75% survival of 2.9 years (95% CI, 2.1e4.4

4

Figure 2. A, Graph showing the probability of first flare after initiation of treatment with a biologic agent. Disease flare occurred in 42.3% of eyes (n ¼ 66). Median time to first flare for the entire cohort was 5.4 years (95% confidence interval [CI], 2.16e5.41 years). B, Graph showing the probability of treatment failure with a biologic agent. Treatment failed in 23.7% of eyes (n ¼ 37). Among those in which treatment failed, the median time to failure was 1.7 years (95% CI, 1.6e1.86 years).

years). Cause of treatment failure was lack of efficacy in 23 eyes and systemic side effects in 5 patients (10 eyes). For eyes in which treatment failed because of lack of efficacy, the 5-year survival rate was 77.8%. Three patients (6 eyes) experienced an allergic reaction to the treatment (1 receiving adalimumab and 2 receiving infliximab), 1 patient (2 eyes) receiving infliximab demonstrated abnormal liver function, and 1 patient (2 eyes) treated with adalimumab experienced unmasking of a demyelinating disease after which treatment was stopped. There were no reported deaths during the follow-up period, but 1 patient was hospitalized and treated for a systemic infection. Nine eyes treated with adalimumab failed at an average time of 2.60.5 years, and 24 eyes treated with infliximab (31.6%) failed at an average time of 1.80.2 years. Baseline data for eyes treated with infliximab and adalimumab demonstrated no significant differences (Table 3). Risk factor analysis for treatment failure revealed that treatment with infliximab, active systemic disease at baseline, and age at baseline were borderline significant risk factors for treatment failure in univariate analysis (Table 4). However, when analyzed together in

Al-Janabi et al



Biologics and Noninfectious Uveitis

Table 3. Demographics of Eyes Treated with Infliximab versus Those Treated with Adalimumab All Eyes Eyes, no. (%) Female gender, no. (%) Age at baseline (yrs) Uveitis type, no. (%) Intermediate Posterior Panuveitis Length of follow-up (yrs) BCVA at baseline (logMAR) Prednisolone dose at baseline (mg)

Infliximab

P Adalimumab Value

136 (100) 76 (55.9) 60 (40.1) 54(39.7) 28 (36.8) 26 (43.3) 0.48 36.111.79 35.032.43 37.4714.86 0.5 0.08 33 (24.3) 14 (18.4) 19 (31.7) 22 (16.2) 16 (21.1) 6 (10) 81 (59.6) 46 (60.5) 35 (58.3) 8.250.45 8.910.61 7.420.65 0.1 0.440.05

0.440.07

0.440.07

16.481.28 15.591.66 17.72.02

1.0 0.42

BCVA ¼ best corrected visual acuity; logMAR ¼ logarithm of the minimum angle of resolution. Data are mean  standard error of mean unless otherwise indicated.

a multivariate model, eyes treated with adalimumab showed an odds ratio for treatment failure of 0.4 (95% CI, 0.2e0.9; P ¼ 0.03) and eyes with concomitantly active systemic disease at baseline showed an odds ratio of 3.2 (95% CI, 1.5e7.1; P ¼ 0.004). For eyes of patients in whom treatment failed with a first-line biologic drug, 38 were switched to an alternative biologic agent: 22 to adalimumab, 6 to infliximab, 6 to etanercept, 2 to tocilizumab, and 2 to golimumab. Of these eyes, treatment failed in 6 eyes treated with etanercept (100%), in 2 eyes treated with golimumab (100%), in 2 eyes treated with tocilizumab (100%), in 5 eyes treated with infliximab (83.3%), and in 2 eyes treated with adalimumab (9.1%). The number of eyes treated with other drugs was not large enough to analyze statistically, but failure rates were 5 eyes (35.7%) among those treated with rituximab at an average time of 1.70.8 years and 4 eyes treated with etanercept (100%) at an average time of 0.50.1 years. Five eyes treated with rituximab (35.7%) at an average of 0.90.4 years and 4 eyes treated with etanercept (100%) failed at an average time of 1.60.2 years. Two eyes of the same patient were treated with vedolizumab for intermediate uveitis-related to ulcerative colitis, and by the end of follow-up (16 months), neither eye had experienced a flare of intermediate uveitis and treatment was not stopped.

Eyes with Behçet’s disease (BD) are known to demonstrate a good response to biologics drugs; therefore, we attempted to examine any differences in response to treatment between eyes with a diagnosis of Behçet’s disease and other causes. There was no difference in change in BCVA, prednisolone dose, or rate of treatment failure. However, there was a smaller rate of flares among eyes with Behçet’s disease compared with other causes (15.7% vs. 55.2%, respectively; P < 0.0001), and the average time to first flare was 9.00.6 months versus 5.50.7 months, respectively (P < 0.0001).

Discussion This study focuses on the use of biologic drugs for the treatment of refractory NIPPU. Although not all biologics show an effect on uveitis, infliximab, adalimumab, and possibly rituximab resulted in rapid control of the inflammation, allowing for a reduction in corticosteroid doses. Treatment failure occurred in less than one quarter of patients and was less likely if the systemic disease was not also active at the time treatment was started or if adalimumab was used. The goal of treating uveitis is to maintain vision and to prevent any deterioration to vision loss.13 Any potential new drug is measured by its success in achieving and maintaining disease control and vision stability. Although corticosteroids remain the cornerstone of immunosuppression, biologic drugs are altering the way we manage uveitis, offering lasting disease control and vision stability. The VISUAL studies demonstrated that treatment with adalimumab delayed the occurrence of disease relapse and paved the way to its licensing for uveitis.10e12 Cohort studies suggested similar effects for other biologic agents, and their offlabel use is growing.14e17 In our cohort, treatment failure with biologics occurred in fewer than one quarter of patients, supporting their role in the management of refractory noninfectious uveitis. In accordance with other reports, etanercept failed to control the ocular inflammation in most patients.18,19 Although physicians advocate its use in systemic immunemediated diseases, reports suggest it is ineffective for treating uveitis and may even increase the risk of uveitis developing in patients with no previous ocular disease.20 Interestingly, rituximab resulted in good inflammatory response, with only one third of eyes failing treatment. Rituximab is well established for the treatment of external

Table 4. Baseline Risk Factors for Treatment Failure among Eyes Treated with Infliximab or Adalimumab Crude Odds Ratio (95% Confidence Interval), P Value Male gender Laterality Anatomic type of uveitis* Posterior Panuveitis Treatment with adalimumab Age at beginning of biologics (yrs) Active systemic disease at baseline Treatment with second-line immunosuppression

Refined Odds Ratio (95% Confidence Interval), P Value

0.6 (0.3e1.2), 0.6 1.1 (0.56e2.18), 0.78 0.75 0.71 0.5 0.98 1.91 0.97

(0.33e1.69), (0.24e2.07), (0.23e1.06), (0.96e1.00), (0.96e3.81), (0.44e2.16),

0.48 0.53 0.07 0.09 0.07 0.94

0.4 (0.18e0.89), 0.03 0.97 (0.95e0.996), 0.03 3.22 (1.46e7.07), 0.004

*Reference set as intermediate uveitis.

5

Ophthalmology Volume -, Number -, Month 2019 ocular and adnexal inflammation and is used for refractory vasculitis-related scleritis, peripheral ulcerative keratitis, or ocular cicatricial pemphigoid. Our results support other studies demonstrating its role in the management of children with juvenile idiopathic arthritis-related uveitis21 or cases of refractory posterior uveitides.14 Although our cohort of patients given rituximab was small, the results support further investigations of this agent as a potential treatment for treatment-resistant uveitis. Use of biologics results in rapid control of the uveitis, and patients subsequently can reduce their dependence on other immunosuppressive drugs.12,22 Although some patients stop using all other medications, continuing treatment with at least a low dose of another immunosuppressive drug can prolong the effect of the biologic drug, even in fully humanized agents like adalimumab.23,24 Most patients in whom treatment fails seem to experience this outcome within the first few months, so that ophthalmologists can be more confident of long-term control after patients have remained quiescent for the first year.25,26 Our results support that adalimumab resulted in a shorter time to response, fewer eyes failing treatment, a longer interval to first disease flare, and a longer time to failure compared with other biologic agents. These failure rates also related to when adalimumab was used as a second-line biologic, suggesting that in those patients in whom another agent failed, switching to adalimumab should be considered. Alternating between agents can maintain long-term disease control and may confer better visual stability. When the ocular inflammation remains active, switching to an alternative biologic agent can re-establish disease control.27e29 The objective of treating uveitis primarily is to preserve vision, and long-term studies indicate that this is possible through strict immunosuppression and addressing any sign of inflammation or ocular complications.1,7,30 This cohort of patients demonstrated a similar trend, with average visual acuity remaining stable throughout follow-up. However, the risk of vision loss developing as a result of ocular complications was almost double that of the general uveitis population,1,31 reflecting more severe, long-lasting ocular inflammation and mainly related to macular pathologic features. The results of the VISUAL III study were published recently and demonstrate that treatment with adalimumab increases the likelihood of disease quiescence, improves visual acuity, and reduces corticosteroid use.12 Prompt treatment, with the aim of achieving complete inflammatory control and early switching of agents when treatment fails, could result in preventing development of vision loss. The main limitation of this study is its retrospective nature and small cohort for some of the agents. However, because uveitis is relatively uncommon and, in most cases, is controlled using corticosteroids and nonbiologic secondline agents, long-term data regarding the use of these drugs are sparse. Treatment with corticosteroids and a second-line agent had failed already in all the eyes in this study; therefore, we did not compare the effect of biologic treatment with these other agents but focused instead on any differences between biologic agents. The results of this study support the continued use and investigation of biologic agents in refractory uveitis, including any influence

6

on side effects related to corticosteroids and secondline immunosuppression drugs. The recent licensing of adalimumab for uveitis and increasing use of these drugs require additional studies to examine their effect and to provide treatment guidelines. In conclusion, this study supports the continued use of biologic agents and particularly antietumor necrosis factor a inhibitors, for the control of refractory noninfectious uveitis. This can result in improved disease control and can prevent disease flares and the development of ocular complications. Expanding the volume of knowledge regarding these drugs will assist ophthalmologists in treating these patients and preserving vision.

References 1. Tomkins-Netzer O, Talat L, Bar A, et al. Long-term clinical outcome and causes of vision loss in patients with uveitis. Ophthalmology. 2014;121(12):2387e2392. 2. Tomkins-Netzer O, Taylor SR, Lightman S. Corticosteroidsparing agents: new treatment options. Dev Ophthalmol. 2012;51:47e56. 3. Jabs DA. Immunosuppression for the uveitides. Ophthalmology. 2018;125(2):193e202. 4. Smith JR, Levinson RD, Holland GN, et al. Differential efficacy of tumor necrosis factor inhibition in the management of inflammatory eye disease and associated rheumatic disease. Arthritis Rheum. 2001;45(3):252e257. 5. Lerman MA, Burnham JM, Chang PY, et al. Response of pediatric uveitis to tumor necrosis factor-alpha inhibitors. J Rheumatol. 2013;40(8):1394e1403. 6. Fabiani C, Vitale A, Rigante D, et al. Comparative efficacy between adalimumab and infliximab in the treatment of noninfectious intermediate uveitis, posterior uveitis, and panuveitis: a retrospective observational study of 107 patients. Clin Rheumatol. 2019;38(2):407e415. 7. Lejoyeux R, Diwo E, Vallet H, et al. Infliximab and adalimumab in uveitic macular edema. Ocul Immunol Inflamm. 2018;26(7):991e996. 8. Fabiani C, Vitale A, Emmi G, et al. Long-term retention rates of adalimumab and infliximab in non-infectious intermediate, posterior, and panuveitis. Clin Rheumatol. 2019;38(1):63e70. 9. Vallet H, Seve P, Biard L, et al. Infliximab versus adalimumab in the treatment of refractory inflammatory uveitis: a multicenter study from the French Uveitis Network. Arthritis Rheumatol. 2016;68(6):1522e1530. 10. Jaffe GJ, Dick AD, Brezin AP, et al. Adalimumab in patients with active noninfectious uveitis. N Engl J Med. 2016;375(10): 932e943. 11. Nguyen QD, Merrill PT, Jaffe GJ, et al. Adalimumab for prevention of uveitic flare in patients with inactive noninfectious uveitis controlled by corticosteroids (VISUAL II): a multicentre, double-masked, randomised, placebo-controlled phase 3 trial. Lancet. 2016;388(10050):1183e1192. 12. Suhler EB, Adan A, Brezin AP, et al. Safety and efficacy of adalimumab in patients with noninfectious uveitis in an ongoing open-label study: VISUAL III. Ophthalmology. 2018;125(7):1075e1087. 13. Niederer RL, Sharief L, Bar A, et al. Predictors of long-term visual outcome in intermediate uveitis. Ophthalmology. 2017;124(3):393e398.

Al-Janabi et al



Biologics and Noninfectious Uveitis

14. Lasave AF, You C, Ma L, et al. Long-term outcomes of rituximab therapy in patients with noninfectious posterior uveitis refractory to conventional immunosuppressive therapy. Retina. 2018;38(2):395e402. 15. Joshi L, Tanna A, McAdoo SP, et al. Long-term outcomes of rituximab therapy in ocular granulomatosis with polyangiitis: impact on localized and nonlocalized disease. Ophthalmology. 2015;122(6):1262e1268. 16. Cordero-Coma M, Calvo-Rio V, Adan A, et al. Golimumab as rescue therapy for refractory immune-mediated uveitis: a threecenter experience. Mediators Inflamm. 2014;2014:717598. 17. Mesquida M, Molins B, Llorenc V, et al. Twenty-four month follow-up of tocilizumab therapy for refractory uveitis-related macular edema. Retina. 2018;38(7):1361e1370. 18. Lim LL, Fraunfelder FW, Rosenbaum JT. Do tumor necrosis factor inhibitors cause uveitis? A registry-based study. Arthritis Rheum. 2007;56(10):3248e3252. 19. Wendling D, Paccou J, Berthelot JM, et al. New onset of uveitis during anti-tumor necrosis factor treatment for rheumatic diseases. Semin Arthritis Rheum. 2011;41(3):503e510. 20. Levy-Clarke G, Jabs DA, Read RW, et al. Expert panel recommendations for the use of anti-tumor necrosis factor biologic agents in patients with ocular inflammatory disorders. Ophthalmology. 2014;121(3):785e796 e3. 21. Miserocchi E, Modorati G, Berchicci L, et al. Long-term treatment with rituximab in severe juvenile idiopathic arthritisassociated uveitis. Br J Ophthalmol. 2016;100(6):782e786. 22. Sharma SM, Damato E, Hinchcliffe AE, et al. Long-term efficacy and tolerability of TNF-a inhibitors in the treatment of non-infectious ocular inflammation: an 8-year prospective surveillance study. Br J Ophthalmol. 2019 Mar 12. https:// doi.org/10.1136/bjophthalmol-2018-312767 [Epub ahead of print].

23. Mercier AE, Ribeiro E, Korobelnik JF, et al. Efficacy of antiTNF-alpha therapy for the treatment of non-infectious uveitis: a retrospective study of 21 patients. Ocul Immunol Inflamm. 2018;26(3):477e484. 24. Riancho-Zarrabeitia L, Calvo-Rio V, Blanco R, et al. AntiTNF-alpha therapy in refractory uveitis associated with sarcoidosis: multicenter study of 17 patients. Semin Arthritis Rheum. 2015;45(3):361e368. 25. Durrani K, Kempen JH, Ying GS, et al. Adalimumab for ocular inflammation. Ocul Immunol Inflamm. 2017;25(3): 405e412. 26. Diaz-Llopis M, Salom D, Garcia-de-Vicuna C, et al. Treatment of refractory uveitis with adalimumab: a prospective multicenter study of 131 patients. Ophthalmology. 2012;119(8): 1575e1581. 27. Dobner BC, Max R, Becker MD, et al. A three-centre experience with adalimumab for the treatment of non-infectious uveitis. Br J Ophthalmol. 2013;97(2):134e138. 28. Dhingra N, Morgan J, Dick AD. Switching biologic agents for uveitis. Eye (Lond). 2009;23(9):1868e1870. 29. Simonini G, Katie D, Cimaz R, et al. Does switching antiTNFalpha biologic agents represent an effective option in childhood chronic uveitis: the evidence from a systematic review and meta-analysis approach. Semin Arthritis Rheum. 2014;44(1):39e46. 30. Tomkins-Netzer O, Lightman S, Drye L, et al. Outcome of treatment of uveitic macular edema: the Multicenter Uveitis Steroid Treatment Trial 2-year results. Ophthalmology. 2015;122(11):2351e2359. 31. Dick AD, Tundia N, Sorg R, et al. Risk of ocular complications in patients with noninfectious intermediate uveitis, posterior uveitis, or panuveitis. Ophthalmology. 2016;123(3): 655e662.

Footnotes and Financial Disclosures Originally received: March 11, 2019. Final revision: July 30, 2019. Accepted: August 28, 2019. Available online: ---.

Author Contributions: Conception and design: Al-Janabi, El Nokrashy, Sharief, Lightman, Tomkins-Netzer Manuscript no. 2019-545.

1

Moorfields Eye Hospital, Institute of Ophthalmology, University College of London, London, United Kingdom.

2

Mansoura Ophthalmic Center, Mansoura University, Cairo, Egypt.

3

Technion, Institute of Technology, Bnai Zion Medical Center, Haifa, Israel.

Analysis and interpretation: Al-Janabi, Sharief, Lightman, Tomkins-Netzer Data collection: Al-Janabi, El Nokrashy, Nagendran, Lightman, TomkinsNetzer Obtained funding: Lightman, Tomkins-Netzer Overall responsibility: Al-Janabi, El Nokrashy, Sharief, Nagendran, Lightman, Tomkins-Netzer

Financial Disclosure(s): The author(s) have made the following disclosure(s): S.L.: Consultant e Allergan, GSK, 4Sight, Bayer, AbbVie, Paraxcel; Advisory board e Allergan, Bayer, AbbVie, Santen, GSK. O.T.-N.: Consultant e Allergan; Lecturer e Bayer; Advisory board e AbbVie.

Abbreviations and Acronyms: BCVA ¼ best-corrected visual acuity; BD ¼ Behçet’s disease; CI ¼ confidence interval; logMAR ¼ logarithm of the minimum angle of resolution; NIPPU ¼ noninfectious intermediate uveitis, posterior uveitis, or panuveitis; SEM ¼ standard error of mean; SUN ¼ standardization of uveitis nomenclature.

HUMAN SUBJECTS: Human subjects were included in this study. The study adhered to the Declaration of Helsinki and was conducted following ethical approval for data collection ROAD16039 (Moorfields Eye Hospital). This is a retrospective study using de-identified subject details. Informed consent was not obtained.

Correspondence: Oren Tomkins-Netzer, MD, PhD, Institute of Ophthalmology, University College of London, 162-165 City Road, London EC1V 2PD, United Kingdom. E-mail: [email protected].

No animal subjects were included in this study.

7