Outcomes of Changing Immunosuppressive Therapy after Treatment Failure in Patients with Noninfectious Uveitis Lavnish Joshi, MD, FRCOphth,1,2 Lazha Talat, MBChB, MPH,1,2 Satish Yaganti, MBBS, MSc,3 Sartaj Sandhu, MBBS,4 Simon R. J. Taylor, PhD, FRCOphth,3,5 Denis Wakefield, MD, FRCPA,6 Peter McCluskey, MD, FRANZCO,4 Susan Lightman, PhD, FRCOphth1,2 Purpose: To evaluate the outcomes of changing immunosuppressive therapy for noninfectious uveitis after failure. Design: Retrospective cohort study. Participants: Patients with noninfectious uveitis managed at 2 tertiary uveitis clinics in the United Kingdom and Australia. Methods: Participants with a history of using immunosuppressive therapy were identified in clinics, and notes were reviewed by doctors trained in uveitis therapy. Each treatment episode/course (starting or changing a therapy) was identified, and demographic details, clinical characteristics, drug used (second-line immunosuppressive agent [ISA] or biologicals), and drug doses were obtained. Main Outcome Measures: For each treatment episode, the reasons for changing therapy, corticosteroidsparing effects, and control of inflammation were determined. Results: A total of 147 patients were identified who underwent 309 different treatment episodes. Fifty-five percent of patients eventually required a change in treatment after their first treatment episode/course. Fortyfive episodes involved switching from one ISA to another, with 50% to 100% of these patients achieving “success” (prednisolone 10 mg and sustained control) with the new ISA. A combination of ISAs were used in 53 episodes, with “success” being achieved in 50% to 71% of these patients. Biological agents were used in 45 episodes, the most common one being infliximab, which achieved success in 80% of patients. Conclusions: Our data suggest that control of inflammation can be achieved after switching or combining ISAs. Ophthalmology 2014;121:1119-1124 ª 2014 by the American Academy of Ophthalmology. Supplemental material is available at www.aaojournal.org.
Systemic corticosteroids are the mainstay of treatment for uveitis resistant to local therapy, bilateral disease, and the more severe forms of uveitis. In some patients, corticosteroids are unable to control inflammation at a dose of 10 mg/ day, or preferably 7.5 mg/day or less,1 whereas in others the side effects of long-term treatment are severe and limit the duration of therapy. These patients require additional immunosuppressive agents (ISAs) to control their disease and to minimize the corticosteroid dose they require.1 Such nonbiologic ISAs include T-cell inhibitors (cyclosporine A [CSA] and tacrolimus), antimetabolites (azathioprine [AZA], methotrexate [MTX], and mycophenolate mofetil [MMF]), and alkylating agents (cyclophosphamide and chlorambucil); ISAs have been reported to become ineffective in up to 17% of patients within the first year.2e6 An expert uveitis panel published a comprehensive set of guidelines on the use of ISAs, but there is no consensus on whether to change ISAs or add a biologic in such patients. There are a limited number of reports describing the outcomes of switching or adding an ISA after failure of an ISA.7 The objectives of this study were to determine the outcome of individual immunosuppressive therapies (both 2014 by the American Academy of Ophthalmology Published by Elsevier Inc.
ISAs and biologicals) in producing a corticosteroid-sparing effect after treatment has failed (because of side effects or ineffectiveness) with a previous agent(s) in a large cohort of patients.
Methods A retrospective review of the clinical records of all patients who attended the uveitis clinics of Susan Lightman at Moorfields’ Eye Hospital (London, UK) and Peter McCluskey at St. Vincent’s clinic (Sydney, Australia) for the management of noninfectious uveitis from January 2010 to August 2012 was undertaken. Ethical approval was obtained from the Moorfields’ Eye Hospital research board under the program of research on causes of visual loss in uveitis (LIGS 10201).
Data Collection, Follow-Up, and Outcome Measures Patients who were currently using or had previously used immunosuppressive drugs or biologics were identified. Patient medical records were reviewed, and information from each treatment episode/course (defined as starting or changing an ISA or biologic) was collected. The following information was collected: age; sex; ISSN 0161-6420/14/$ - see front matter http://dx.doi.org/10.1016/j.ophtha.2013.11.032
1119
Ophthalmology Volume 121, Number 5, May 2014 uveitis subtype; any associated medical condition (e.g., sarcoidosis); duration of disease and follow-up; prednisolone dose and disease activity (based on the Standardization of Uveitis Nomenclature workshop grading system)8 at the start of treatment episode; baseline and final best-corrected visual acuity (VA); any decrease/ gain in best-corrected VA by 2 Early Treatment Diabetic Retinopathy Study (ETDRS) lines due to inflammation; ocular complications; maximum dose of ISA; reasons for changing treatment; and corticosteroid-sparing effect of ISA or biologics (defined as time to reach a prednisolone dose 7.5 or 10 mg with maintained inactivity spanning at least 28 days).5 Patients were followed up until their last recorded visit to the clinic. Visual acuity was recorded using Snellen VA charts and converted to approximate ETDRS scores, which are more intuitively interpretable than logarithm of the minimum angle of resolution units.9 Other associated features of uveitis were documented, such as increased vitritis, new-onset macular edema, vasculitis, or optic neuropathy/ swelling. The definition of remission was based on the clinician’s evaluation of the patient, noting that the treatment could be reduced or stopped because there was no disease activity for at least 6 months after starting the agent; this definition also included no relapse within 6 months after stopping the agent. Ineffectiveness was based on the clinician’s impression that a change in therapy was needed because the agent failed to improve inflammation or recurrent relapses occurred while on the treatment regimen.
Statistical Analyses All data were entered into a Microsoft Excel 2007 spreadsheet (Microsoft Corp., Redmond, WA) and analyzed using the Excel PivotTable function and GraphPad Prism v5.01 (GraphPad Software, La Jolla, CA).
Results The demographics and uveitis phenotypes of the patients are detailed in Table 1. We identified a total of 147 patients who underwent a total of 309 different treatment episodes; 30 episodes were a reduction of treatment (reducing from a combination of therapies to a single therapy), 56 episodes were a switch from one systemic therapy to another, 87 episodes were an increase in treatment (increasing from a single therapy to a combination of therapies), and 9 episodes were due to the reintroduction of a treatment that had been stopped. A total of 127 episodes involved the use of a single agent as the first treatment episode.
Continuation and Change of Treatment Figure 1 shows that 55% of patients eventually required a change in treatment after their first treatment episode/course. Some patients subsequently underwent up to 8 treatment courses/episodes, highlighting that is difficult to achieve remission in these patients with a single course of treatment.
Visual Acuity Changes across Treatment Courses Figure 2 shows the last recorded ETDRS-equivalent VA in all eyes for each consecutive treatment course and reveals a decline in VA after consecutive treatment courses. There may be a trend for patients to lose vision with increasing courses of treatment. However, up to 22% of patients gained 2 or more ETDRS lines of VA after treatment (Fig 3). The causes of visual loss are shown in Table 2.
Effectiveness of Single Immunosuppressive Agents after Failure of Other Agents in Comparison with Their Use as Initial Agents Of the 127 treatment episodes involving the use of a single agent during the first treatment episode, MMF (44%) was the most commonly used initial agent, followed by CSA (25%), AZA (16%), MTX (13%), and alkylating agents (2%). The mean doses used for each of these initial ISAs are summarized in Table 3 (available at www.aaojournal.org). However, these agents failed (stopped because of ineffectiveness or side effects) in 40% to 75% of patients (Table 4), with the highest proportion for those given CSA. Remission (stopping treatment because of maintained inactivity) was not achieved in many patients, but a greater proportion of patients taking AZA managed to achieve this (35%). A greater proportion of patients taking AZA had systemic disease and a longer duration of disease and posterior uveitis than those taking the other ISAs (Table 1). A statistical comparison of baseline factors between ISAs was not possible because it would violate the rule of independent samples (because some patients may have multiple ISAs). Of the 56 episodes that involved switching from one therapy to another, 45 involved switching from one ISA to another, whereas the other 9 involved switching between biological agents or switching an agent in an ISA combination regimen. Table 3 (available at www.aaojournal.org) shows the summary and outcomes of ISAs used after the failure of other ISAs, comparing them with ISAs used as initial therapy. The most commonly used ISAs after ISA failure were MTX (18 episodes) and MMF
Table 1. Baseline Characteristics of All Patients and First Treatment Episode for Commonly Used Immunosuppressive Agents
Characteristic Age (yrs), median (range) Female sex (%) Site of inflammation (%) Anterior uveitis Intermediate uveitis Posterior uveitis Systemic disease Duration of disease (mos), median (range) Ocular complications (%)
Initial Agent Used (No. of Episodes)
All Patients (n [ 147)
MTX (n ¼ 16)
AZA (n ¼ 20)
MMF (n ¼ 56)
CSA (n ¼ 32)
37 (5e75) 55
31 (10e59) 69
37 (20e64) 40
42 (5e75) 43
33 (17e54) 34
16 20 64 53 31 (0e528) 67
54 13 33 63 38 (0e156) 63
10.5 10.5 79 95 74 (0e364) 60
5.5 32 62.5 38 46 (0e249) 73
10 13 77 63 20 (0e95) 69
AZA ¼ azathioprine; CSA ¼ cyclosporine A; MMF ¼ mycophenolate mofetil; MTX ¼ methotrexate.
1120
Joshi et al
Outcomes of Changing Immunosuppression in Uveitis
Figure 3. Proportion of patients gaining approximately 2 Early Treatment Diabetic Retinopathy Study (ETDRS) visual acuity lines across consecutive treatment courses. SEM ¼ standard error of mean.
Figure 1. Continuation and change of therapy across consecutive treatment courses. Asterisk indicates proportion of total cohort studied. Bars indicate treatment status (stopped because of remission, changed, continuation) for each course. Numbers within bars indicate the proportion of patients in each treatment course and their treatment status.
(16 episodes), with success (prednisolone dose 10 mg/day) still being possible in 63% and 64%, respectively. Furthermore, a gain in vision (2 ETDRS lines) was still possible in 33% of patients switching to MTX and 44% of patients switching to MMF. This suggests that these agents can still be effective in controlling disease, even if they are used after other ISAs have failed. However, some of these patients still went on to have disease relapse, and a decline in vision (2 ETDRS lines due to disease relapse) at any point was still experienced in 31% to 100% of patients after they switched to an ISA, highlighting that long-term inactivity is difficult to achieve.
Commonly Used Immunosuppressive Agent Combinations and Outcomes Of the 87 episodes in which a combination of drugs was used, ISA combinations were used in 53 courses, and 34 courses used biological agents along with another ISA.
The most commonly used ISA combinations are summarized in Table 5. These included AZA þ CSA (14 episodes), CSA þ MMF (20 episodes), and CSA þ MTX (4 episodes). Success at a prednisolone dose of 10 mg/day was achieved in 50% to 71% of these patients; a gain in VA (2 ETDRS lines) was achieved in 21% to 30% of patients. However, 21% to 31% of these patients also experienced subsequent disease relapses with a decline in vision (2 ETDRS lines) at any point after these ISA combinations were used. These ISA combinations were stopped in up to 50% of patients because of ineffectiveness and in up to 29% because of side effects. During these episodes, the mean (standard deviation) doses for MMF, MTX, AZA, and CSA were 1.9 g/day (0.5), 17 mg/week (6.7), 155 mg/day (81), and 303 mg/day (158), respectively.
Commonly Used Biological Agents and Outcomes Biological agents were used in a total of 45 episodes. The outcomes of the most commonly used biological agents are summarized in Table 6. These included infliximab (30 episodes) and adalimumab (8 episodes). The main reason for using a biological agent was ineffectiveness of other ISAs in 94%. These biological agents were able to achieve high success rates, but patients still experienced a decline in vision (2 ETDRS lines) because of inflammation at any point while receiving these therapies. Infliximab was stopped in 17% of patients because of ineffectiveness, whereas adalimumab was stopped in 63% for the same reason.
Discussion This study demonstrates that a large proportion (60%) of patients with uveitis undergo more than 1 change in their Table 2. Causes of Visual Loss
Figure 2. Visual acuity (VA) changes across consecutive treatment courses (based on final VA for each course). Approximate Early Treatment Diabetic Retinopathy Study (ETDRS) letter score. SEM ¼ standard error of mean.
Causes of Visual Loss
Patients (n [ 147)
Cataract Glaucoma Ischemia/artery occlusion/vein occlusion Epiretinal membrane/macular hole Neovascular membrane CME/vitritis
33% 14% 15% 10% 6% 40%
CME ¼ cystoid macular edema.
1121
Ophthalmology Volume 121, Number 5, May 2014 Table 4. Reasons for Discontinuation of Initial Immunosuppressive Agent Reason for Discontinuation (%) Initial Agent
Not Side Stopped (%) Ineffective Effects Remission Unknown
MMF (n ¼ 56) MTX (n ¼ 16) AZA (n ¼ 20) CSA (n ¼ 32) ALK (n ¼ 2)
23 31 20 13 0
50 56 35 66 50
7 0 5 9 0
18 13 35 9 50
2 0 5 3 0
ALK ¼ alkylating agents; AZA ¼ azathioprine; CSA ¼ cyclosporine A; MMF ¼ mycophenolate mofetil; MTX ¼ methotrexate.
immunosuppressive treatment with ISAs or biological agents with time. Subsequent changes in treatment can be effective, including switching to another ISA after failure of one or adding another ISA to the treatment regimen. Biological agents were not frequently used, largely because of the difficulty in obtaining patient funding, but infliximab had a good rate of success. There seems to be a trend in a reduction in vision across consecutive treatment courses, but a gain in vision is still possible in 12% to 22% of patients after changing ISAs. The use of any antimetabolites after the failure of other immunosuppressive regimens led to success (prednisolone 10 mg/day) in 48% to 60% of cases. The use of CSA after failure of other regimens also led to success in some cases. Using a combination of ISAs was also useful and, depending on the combination used, led to success in 50% to 71% of cases. It is possible that a combination of agents with a different mechanism of action could have a synergistic effect in controlling inflammation. The effectiveness of ISA combinations in uveitis has not been widely reported in the literature, but this strategy is used by uveitis specialists. Menezo et al10 reported that 32% of their patients required the addition of a third agent (prednisolone and 2 ISAs). In the Systemic Immunosuppressive Therapy for Eye Diseases (SITE) cohort study, the proportion of patients requiring a third agent was 16%, 17%, 9%, and 19% in the MTX, CSA, AZA, and MMF groups, respectively.2e5 The efficacy and tolerance of combination strategies have been evaluated in only a few studies. One of the earliest studies found that a combination of steroids, MTX, and CSA (5 mg/kg/day) was able to achieve “total and lasting remission” in all 32 patients with endogenous noninfectious
uveitis, with the only side effect being hirsutism in 2 cases.11 They suggested that the combination therapy was able to achieve control of inflammation with lower doses of each agent than monotherapy would allow, thus reducing the toxicity of each individual agent. More recently, Kiss et al12 reported that several strategies, including the use of CSA and MMF, were able to preserve visual function in patients with birdshot retinochoroidopathy. A previous report of patients with the same condition found no difference in the rate of visual loss between untreated patients and patients treated with a single second-line ISA.13 Sobrin et al7 had to add CSA to MMF in 9 of 85 patients (11%) with ocular inflammatory disease who failed MTX, and all these patients having persistent inflammation despite the addition of a second agent. The success in controlling ocular inflammation after switching from one ISA to another has not been widely reported. Sobrin et al7 found that switching to MMF after MTX failure or intolerance is effective in controlling inflammation in 55% of patients with uveitis/scleritis. In the SITE cohort study, patients switching from T-cell inhibitors to MTX were 3 times more likely to gain control of inflammation than patients using MTX for the first time, but the prior use of other ISAs was not associated with significantly different rates of treatment success.2 This suggests that patients who have failed other agents may still respond to MTX. This is in contrast to our findings because a lower proportion of patients achieved success (50%) when switched to MTX in comparison with when it was used as the initial agent. In our study, switching to MMF achieved success in a greater proportion of patients than when it was used as the initial agent, which is again in contrast with the SITE cohort.4 It is possible that the higher MMF dose used in patients who had previously failed ISAs contributed to this difference. Our study did not use the maximum doses of MMF and MTX given by the Uveitis Expert Panel guidelines1 because increasing doses beyond the average doses used in our study led to increased side effects with a limited increase in efficacy. In patients switching to AZA, those previously taking other antimetabolites or Tcell inhibitors tended to respond to AZA less often than patients not previously taking these drugs,3 which is in contrast to our findings, suggesting that AZA was able to achieve control (inactivity at prednisolone 10 mg) in a similar proportion of patients when used as an initial agent or when switching to it after other agents (Table 3, available at www.aaojournal.org). The prior use of other
Table 5. Outcomes of Commonly Used Combination Agents Success (%)
VA Changes (%)
Reason for Discontinuation (%)
Agents Used
Prednisolone 10 mg
Prednisolone 7.5 mg
Gain 2 Lines
Loss 2 Lines
Ineffective
Side Effects
Remission
AZA þ CSA (n ¼ 14) CSA þ MMF (n ¼ 20) CSA þ MTX (n ¼ 4)
71 65 50
29 25 0
21 30 0
21 35 25
29 50 50
29 15 25
36 10 25
ALK ¼ alkylating agents; AZA ¼ azathioprine; CSA ¼ cyclosporine A; MMF ¼ mycophenolate mofetil; MTX ¼ methotrexate; VA ¼ visual acuity.
1122
Joshi et al
Outcomes of Changing Immunosuppression in Uveitis
Table 6. Outcomes of Biological Agents after Failure of Other Agents Success (%)
VA Changes (%)
Reason for Discontinuation (%)
Agent Used
Prednisolone 10 mg
Prednisolone 7.5 mg
Gain 2 Lines
Loss 2 Lines
Ineffective
Side Effects
Remission
Unknown
Infliximab (n ¼ 30) Adalimumab (n ¼ 8)
80 88
60 75
17 25
20 50
17 63
13 0
23 0
7 0
VA ¼ visual acuity.
ISAs in patients switching to CSA was not associated with differences in subsequent treatment success compared with those patients using CSA for the first time in the SITE cohort.5 The definition of success used in our study was similar to that used by the SITE cohort study. During their sensitivity analysis, the authors found that not having such a stringent definition of success (i.e., success to be sustained) also resulted in a larger proportion of patients achieving success on the basis of this definition. However, their analyses found that for some ISAs, 83% of patients who achieved control would subsequently have had a relapse by the next visit,2 suggesting that many patients relapse just after achieving control of inflammation or corticosteroid-sparing goals. Local and topical steroids were increased for many of the patients in our study, without systemic steroids being increased. Indications included an increase in anterior chamber cell activity (topical steroids) and the presence of cystoid macular edema (CME) and vitritis (local steroids). In the SITE cohort study, patients may have achieved quiescence if there was no cellular activity and prednisolone doses 10 mg, but there is no indication of whether these patients required an increase in local steroids to treat CME. If there were such patients, did they truly achieve “sustained control” of inflammation? Thus, such patients would actually overestimate the benefits of treatment effect. Whether CME is a measure of disease activity is a matter of debate. Cystoid macular edema certainly can be treated in a similar way as ocular cellular activity through the increase of oral corticosteroids,14-16 which may actually lead to a more rapid decrease of CME than with periocular steroid injections.17 However, the SUN group regards CME as a structural complication of uveitis,8 rather than a measure of active disease. The basis of this assumption shares common ground with the explanation for the presence of anterior chamber flare in chronic uveitis. The presence of CME and anterior chamber flare is due to vascular damage from the inflammatory process, and both may be long-standing, if not permanent; thus, both are believed to be a marker of past inflammation but not a sign of active inflammation.18,19
Study Limitations The limitations of our study include its retrospective nature, which can lead to imprecise and nonstandardized data collection from assessment at irregular intervals and differential follow-up. Treatment bias may have existed because treatment was not randomized or administered in a standard manner.
The SITE cohort study did not report VA outcomes, which can be difficult to report in a standardized manner. Several approaches can be taken. In studies in which the follow-up time is variable, the proportion with a particular threshold VA at presentation and rates of VA events during follow-up (e.g., falling below a specified threshold because of disease activity) can be reported,8,20 which is the method we adopted in our study. However, this can be problematic because a decline in VA may not be a direct result of the current ISA regimen that a patient is receiving. For example, a cataract may develop in a patient receiving a particular ISA, but although this ISA may have achieved sustained control for this patient and the decline may not be due to disease activity, any improvement in VA caused by the ISA regimen would be limited by the cataract.
References 1. Jabs DA, Rosenbaum JT, Foster CS, et al. Guidelines for the use of immunosuppressive drugs in patients with ocular inflammatory disorders: recommendations of an expert panel. Am J Ophthalmol 2000;130:492–513. 2. Gangaputra S, Newcomb CW, Liesegang TL, et al. Systemic Immunosuppressive Therapy for Eye Diseases Cohort Study. Methotrexate for ocular inflammatory diseases. Ophthalmology 2009;116:2188–98. 3. Pasadhika S, Kempen JH, Newcomb CW, et al. Azathioprine for ocular inflammatory diseases. Am J Ophthalmol 2009;148: 500–9. 4. Daniel E, Thorne JE, Newcomb CW, et al. Mycophenolate mofetil for ocular inflammation. Am J Ophthalmol 2010;149:423–32. 5. Kacmaz RO, Kempen JH, Newcomb C, et al. Cyclosporine for ocular inflammatory diseases. Ophthalmology 2010;117:576–84. 6. Pujari SS, Kempen JH, Newcomb CW, et al. Cyclophosphamide for ocular inflammatory diseases. Ophthalmology 2010;117:356–65. 7. Sobrin L, Christen W, Foster CS. Mycophenolate mofetil after methotrexate failure or intolerance in the treatment of scleritis and uveitis. Ophthalmology 2008;115:1416–21. 8. Standardization of Uveitis Nomenclature (SUN) Working Group. Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop. Am J Ophthalmol 2005;140:509–16. 9. Gregori NZ, Feuer W, Rosenfeld PJ. Novel method for analyzing Snellen visual acuity measurements. Retina 2010;30:1046–50. 10. Menezo V, Lau C, Comer M, Lightman S. Clinical outcome of chronic immunosuppression in patients with non-infectious uveitis. Clin Experiment Ophthalmol 2005;33:16–21.
1123
Ophthalmology Volume 121, Number 5, May 2014 11. Pascalis L, Pia G, Aresu G, et al. Combined cyclosporin Asteroid-MTX treatment in endogenous non-infectious uveitis. J Autoimmun 1993;6:467–80. 12. Kiss S, Ahmed M, Letko E, Foster CS. Long-term follow-up of patients with birdshot retinochoroidopathy treated with corticosteroid-sparing systemic immunomodulatory therapy. Ophthalmology 2005;112:1066–71. 13. Rothova A, Berendschot TT, Probst K, et al. Birdshot chorioretinopathy: long-term manifestations and visual prognosis. Ophthalmology 2004;111:954–9. 14. Rothova A. Inflammatory cystoid macular edema. Curr Opin Ophthalmol 2007;18:487–92. 15. McCluskey PJ, Towler HM, Lightman S. Management of chronic uveitis. BMJ 2000;320:555–8.
16. Okhravi N, Lightman S. Cystoid macular edema in uveitis. Ocul Immunol Inflamm 2003;11:29–38. 17. Venkatesh P, Abhas Z, Garg S, Vohra R. Prospective optical coherence tomographic evaluation of the efficacy of oral and posterior subtenon corticosteroids in patients with intermediate uveitis. Graefes Arch Clin Exp Ophthalmol 2007;245:59–67. 18. Samson CM, Waheed N, Baltatzis S, Foster CS. Methotrexate therapy for chronic noninfectious uveitis: analysis of a case series of 160 patients. Ophthalmology 2001;108:1134–9. 19. Chorich LJ III, Klisovic DD, Foster CS. Diagnosis of uveitis. In: Foster CS, Vitale AT. Diagnosis and Treatment of Uveitis. 2nd ed. New Delhi: Jaypee Brothers Medical; 2013:101–30. 20. Jabs DA. Improving the reporting of clinical case series. Am J Ophthalmol 2005;139:900–5.
Footnotes and Financial Disclosures Originally received: June 11, 2013. Final revision: November 16, 2013. Accepted: November 18, 2013. Available online: January 14, 2014.
Manuscript no. 2013-933.
1
Moorfields Eye Hospital, London, United Kingdom.
2
UCL Institute of Ophthalmology, London, United Kingdom.
3
Royal Surrey County Hospital, Guildford, United Kingdom.
4
Save Sight Institute, Sydney Eye Hospital, Sydney, Australia.
5
Faculty of Medicine, Imperial College London, Hammersmith Hospital, London, United Kingdom.
6
School of Medical Sciences, University of New South Wales, Kensington, New South Wales, Australia.
1124
Financial Disclosure(s): The author(s) have made the following disclosure(s): S.R.J.T. has board membership with Allergan and has received financial support for the development of educational presentations by Allergan. S.L. has board membership with Allergan and GlaxoSmithKline. Funding: L.J. and S.R.J.T. were supported by the UK National Institute of Health Research. S.L. has received consultancy fees from PAREXEL International Corp., Bayer, and Zeiss, and received payment for lectures and development of educational materials from Allergan. Correspondence: Susan Lightman, PhD, FRCOphth, UCL Institute of Ophthalmology, 11-43 Bath Street, London, EC1V 9EL. E-mail: [email protected].
Systemic corticosteroids are the mainstay of treatment for uveitis resistant to local therapy, bilateral disease, and the more severe forms of uveitis. In some patients, corticosteroids are unable to control inflammation at a dose of 10 mg/ day, or preferably 7.5 mg/day or less,1 whereas in others the side effects of long-term treatment are severe and limit the duration of therapy. These patients require additional immunosuppressive agents (ISAs) to control their disease and to minimize the corticosteroid dose they require.1 Such nonbiologic ISAs include T-cell inhibitors (cyclosporine A [CSA] and tacrolimus), antimetabolites (azathioprine [AZA], methotrexate [MTX], and mycophenolate mofetil [MMF]), and alkylating agents (cyclophosphamide and chlorambucil); ISAs have been reported to become ineffective in up to 17% of patients within the first year.2e6 An expert uveitis panel published a comprehensive set of guidelines on the use of ISAs, but there is no consensus on whether to change ISAs or add a biologic in such patients. There are a limited number of reports describing the outcomes of switching or adding an ISA after failure of an ISA.7 The objectives of this study were to determine the outcome of individual immunosuppressive therapies (both 2014 by the American Academy of Ophthalmology Published by Elsevier Inc.
ISAs and biologicals) in producing a corticosteroid-sparing effect after treatment has failed (because of side effects or ineffectiveness) with a previous agent(s) in a large cohort of patients.
Methods A retrospective review of the clinical records of all patients who attended the uveitis clinics of Susan Lightman at Moorfields’ Eye Hospital (London, UK) and Peter McCluskey at St. Vincent’s clinic (Sydney, Australia) for the management of noninfectious uveitis from January 2010 to August 2012 was undertaken. Ethical approval was obtained from the Moorfields’ Eye Hospital research board under the program of research on causes of visual loss in uveitis (LIGS 10201).
Data Collection, Follow-Up, and Outcome Measures Patients who were currently using or had previously used immunosuppressive drugs or biologics were identified. Patient medical records were reviewed, and information from each treatment episode/course (defined as starting or changing an ISA or biologic) was collected. The following information was collected: age; sex; ISSN 0161-6420/14/$ - see front matter http://dx.doi.org/10.1016/j.ophtha.2013.11.032
1119
Ophthalmology Volume 121, Number 5, May 2014 uveitis subtype; any associated medical condition (e.g., sarcoidosis); duration of disease and follow-up; prednisolone dose and disease activity (based on the Standardization of Uveitis Nomenclature workshop grading system)8 at the start of treatment episode; baseline and final best-corrected visual acuity (VA); any decrease/ gain in best-corrected VA by 2 Early Treatment Diabetic Retinopathy Study (ETDRS) lines due to inflammation; ocular complications; maximum dose of ISA; reasons for changing treatment; and corticosteroid-sparing effect of ISA or biologics (defined as time to reach a prednisolone dose 7.5 or 10 mg with maintained inactivity spanning at least 28 days).5 Patients were followed up until their last recorded visit to the clinic. Visual acuity was recorded using Snellen VA charts and converted to approximate ETDRS scores, which are more intuitively interpretable than logarithm of the minimum angle of resolution units.9 Other associated features of uveitis were documented, such as increased vitritis, new-onset macular edema, vasculitis, or optic neuropathy/ swelling. The definition of remission was based on the clinician’s evaluation of the patient, noting that the treatment could be reduced or stopped because there was no disease activity for at least 6 months after starting the agent; this definition also included no relapse within 6 months after stopping the agent. Ineffectiveness was based on the clinician’s impression that a change in therapy was needed because the agent failed to improve inflammation or recurrent relapses occurred while on the treatment regimen.
Statistical Analyses All data were entered into a Microsoft Excel 2007 spreadsheet (Microsoft Corp., Redmond, WA) and analyzed using the Excel PivotTable function and GraphPad Prism v5.01 (GraphPad Software, La Jolla, CA).
Results The demographics and uveitis phenotypes of the patients are detailed in Table 1. We identified a total of 147 patients who underwent a total of 309 different treatment episodes; 30 episodes were a reduction of treatment (reducing from a combination of therapies to a single therapy), 56 episodes were a switch from one systemic therapy to another, 87 episodes were an increase in treatment (increasing from a single therapy to a combination of therapies), and 9 episodes were due to the reintroduction of a treatment that had been stopped. A total of 127 episodes involved the use of a single agent as the first treatment episode.
Continuation and Change of Treatment Figure 1 shows that 55% of patients eventually required a change in treatment after their first treatment episode/course. Some patients subsequently underwent up to 8 treatment courses/episodes, highlighting that is difficult to achieve remission in these patients with a single course of treatment.
Visual Acuity Changes across Treatment Courses Figure 2 shows the last recorded ETDRS-equivalent VA in all eyes for each consecutive treatment course and reveals a decline in VA after consecutive treatment courses. There may be a trend for patients to lose vision with increasing courses of treatment. However, up to 22% of patients gained 2 or more ETDRS lines of VA after treatment (Fig 3). The causes of visual loss are shown in Table 2.
Effectiveness of Single Immunosuppressive Agents after Failure of Other Agents in Comparison with Their Use as Initial Agents Of the 127 treatment episodes involving the use of a single agent during the first treatment episode, MMF (44%) was the most commonly used initial agent, followed by CSA (25%), AZA (16%), MTX (13%), and alkylating agents (2%). The mean doses used for each of these initial ISAs are summarized in Table 3 (available at www.aaojournal.org). However, these agents failed (stopped because of ineffectiveness or side effects) in 40% to 75% of patients (Table 4), with the highest proportion for those given CSA. Remission (stopping treatment because of maintained inactivity) was not achieved in many patients, but a greater proportion of patients taking AZA managed to achieve this (35%). A greater proportion of patients taking AZA had systemic disease and a longer duration of disease and posterior uveitis than those taking the other ISAs (Table 1). A statistical comparison of baseline factors between ISAs was not possible because it would violate the rule of independent samples (because some patients may have multiple ISAs). Of the 56 episodes that involved switching from one therapy to another, 45 involved switching from one ISA to another, whereas the other 9 involved switching between biological agents or switching an agent in an ISA combination regimen. Table 3 (available at www.aaojournal.org) shows the summary and outcomes of ISAs used after the failure of other ISAs, comparing them with ISAs used as initial therapy. The most commonly used ISAs after ISA failure were MTX (18 episodes) and MMF
Table 1. Baseline Characteristics of All Patients and First Treatment Episode for Commonly Used Immunosuppressive Agents
Characteristic Age (yrs), median (range) Female sex (%) Site of inflammation (%) Anterior uveitis Intermediate uveitis Posterior uveitis Systemic disease Duration of disease (mos), median (range) Ocular complications (%)
Initial Agent Used (No. of Episodes)
All Patients (n [ 147)
MTX (n ¼ 16)
AZA (n ¼ 20)
MMF (n ¼ 56)
CSA (n ¼ 32)
37 (5e75) 55
31 (10e59) 69
37 (20e64) 40
42 (5e75) 43
33 (17e54) 34
16 20 64 53 31 (0e528) 67
54 13 33 63 38 (0e156) 63
10.5 10.5 79 95 74 (0e364) 60
5.5 32 62.5 38 46 (0e249) 73
10 13 77 63 20 (0e95) 69
AZA ¼ azathioprine; CSA ¼ cyclosporine A; MMF ¼ mycophenolate mofetil; MTX ¼ methotrexate.
1120
Joshi et al
Outcomes of Changing Immunosuppression in Uveitis
Figure 3. Proportion of patients gaining approximately 2 Early Treatment Diabetic Retinopathy Study (ETDRS) visual acuity lines across consecutive treatment courses. SEM ¼ standard error of mean.
Figure 1. Continuation and change of therapy across consecutive treatment courses. Asterisk indicates proportion of total cohort studied. Bars indicate treatment status (stopped because of remission, changed, continuation) for each course. Numbers within bars indicate the proportion of patients in each treatment course and their treatment status.
(16 episodes), with success (prednisolone dose 10 mg/day) still being possible in 63% and 64%, respectively. Furthermore, a gain in vision (2 ETDRS lines) was still possible in 33% of patients switching to MTX and 44% of patients switching to MMF. This suggests that these agents can still be effective in controlling disease, even if they are used after other ISAs have failed. However, some of these patients still went on to have disease relapse, and a decline in vision (2 ETDRS lines due to disease relapse) at any point was still experienced in 31% to 100% of patients after they switched to an ISA, highlighting that long-term inactivity is difficult to achieve.
Commonly Used Immunosuppressive Agent Combinations and Outcomes Of the 87 episodes in which a combination of drugs was used, ISA combinations were used in 53 courses, and 34 courses used biological agents along with another ISA.
The most commonly used ISA combinations are summarized in Table 5. These included AZA þ CSA (14 episodes), CSA þ MMF (20 episodes), and CSA þ MTX (4 episodes). Success at a prednisolone dose of 10 mg/day was achieved in 50% to 71% of these patients; a gain in VA (2 ETDRS lines) was achieved in 21% to 30% of patients. However, 21% to 31% of these patients also experienced subsequent disease relapses with a decline in vision (2 ETDRS lines) at any point after these ISA combinations were used. These ISA combinations were stopped in up to 50% of patients because of ineffectiveness and in up to 29% because of side effects. During these episodes, the mean (standard deviation) doses for MMF, MTX, AZA, and CSA were 1.9 g/day (0.5), 17 mg/week (6.7), 155 mg/day (81), and 303 mg/day (158), respectively.
Commonly Used Biological Agents and Outcomes Biological agents were used in a total of 45 episodes. The outcomes of the most commonly used biological agents are summarized in Table 6. These included infliximab (30 episodes) and adalimumab (8 episodes). The main reason for using a biological agent was ineffectiveness of other ISAs in 94%. These biological agents were able to achieve high success rates, but patients still experienced a decline in vision (2 ETDRS lines) because of inflammation at any point while receiving these therapies. Infliximab was stopped in 17% of patients because of ineffectiveness, whereas adalimumab was stopped in 63% for the same reason.
Discussion This study demonstrates that a large proportion (60%) of patients with uveitis undergo more than 1 change in their Table 2. Causes of Visual Loss
Figure 2. Visual acuity (VA) changes across consecutive treatment courses (based on final VA for each course). Approximate Early Treatment Diabetic Retinopathy Study (ETDRS) letter score. SEM ¼ standard error of mean.
Causes of Visual Loss
Patients (n [ 147)
Cataract Glaucoma Ischemia/artery occlusion/vein occlusion Epiretinal membrane/macular hole Neovascular membrane CME/vitritis
33% 14% 15% 10% 6% 40%
CME ¼ cystoid macular edema.
1121
Ophthalmology Volume 121, Number 5, May 2014 Table 4. Reasons for Discontinuation of Initial Immunosuppressive Agent Reason for Discontinuation (%) Initial Agent
Not Side Stopped (%) Ineffective Effects Remission Unknown
MMF (n ¼ 56) MTX (n ¼ 16) AZA (n ¼ 20) CSA (n ¼ 32) ALK (n ¼ 2)
23 31 20 13 0
50 56 35 66 50
7 0 5 9 0
18 13 35 9 50
2 0 5 3 0
ALK ¼ alkylating agents; AZA ¼ azathioprine; CSA ¼ cyclosporine A; MMF ¼ mycophenolate mofetil; MTX ¼ methotrexate.
immunosuppressive treatment with ISAs or biological agents with time. Subsequent changes in treatment can be effective, including switching to another ISA after failure of one or adding another ISA to the treatment regimen. Biological agents were not frequently used, largely because of the difficulty in obtaining patient funding, but infliximab had a good rate of success. There seems to be a trend in a reduction in vision across consecutive treatment courses, but a gain in vision is still possible in 12% to 22% of patients after changing ISAs. The use of any antimetabolites after the failure of other immunosuppressive regimens led to success (prednisolone 10 mg/day) in 48% to 60% of cases. The use of CSA after failure of other regimens also led to success in some cases. Using a combination of ISAs was also useful and, depending on the combination used, led to success in 50% to 71% of cases. It is possible that a combination of agents with a different mechanism of action could have a synergistic effect in controlling inflammation. The effectiveness of ISA combinations in uveitis has not been widely reported in the literature, but this strategy is used by uveitis specialists. Menezo et al10 reported that 32% of their patients required the addition of a third agent (prednisolone and 2 ISAs). In the Systemic Immunosuppressive Therapy for Eye Diseases (SITE) cohort study, the proportion of patients requiring a third agent was 16%, 17%, 9%, and 19% in the MTX, CSA, AZA, and MMF groups, respectively.2e5 The efficacy and tolerance of combination strategies have been evaluated in only a few studies. One of the earliest studies found that a combination of steroids, MTX, and CSA (5 mg/kg/day) was able to achieve “total and lasting remission” in all 32 patients with endogenous noninfectious
uveitis, with the only side effect being hirsutism in 2 cases.11 They suggested that the combination therapy was able to achieve control of inflammation with lower doses of each agent than monotherapy would allow, thus reducing the toxicity of each individual agent. More recently, Kiss et al12 reported that several strategies, including the use of CSA and MMF, were able to preserve visual function in patients with birdshot retinochoroidopathy. A previous report of patients with the same condition found no difference in the rate of visual loss between untreated patients and patients treated with a single second-line ISA.13 Sobrin et al7 had to add CSA to MMF in 9 of 85 patients (11%) with ocular inflammatory disease who failed MTX, and all these patients having persistent inflammation despite the addition of a second agent. The success in controlling ocular inflammation after switching from one ISA to another has not been widely reported. Sobrin et al7 found that switching to MMF after MTX failure or intolerance is effective in controlling inflammation in 55% of patients with uveitis/scleritis. In the SITE cohort study, patients switching from T-cell inhibitors to MTX were 3 times more likely to gain control of inflammation than patients using MTX for the first time, but the prior use of other ISAs was not associated with significantly different rates of treatment success.2 This suggests that patients who have failed other agents may still respond to MTX. This is in contrast to our findings because a lower proportion of patients achieved success (50%) when switched to MTX in comparison with when it was used as the initial agent. In our study, switching to MMF achieved success in a greater proportion of patients than when it was used as the initial agent, which is again in contrast with the SITE cohort.4 It is possible that the higher MMF dose used in patients who had previously failed ISAs contributed to this difference. Our study did not use the maximum doses of MMF and MTX given by the Uveitis Expert Panel guidelines1 because increasing doses beyond the average doses used in our study led to increased side effects with a limited increase in efficacy. In patients switching to AZA, those previously taking other antimetabolites or Tcell inhibitors tended to respond to AZA less often than patients not previously taking these drugs,3 which is in contrast to our findings, suggesting that AZA was able to achieve control (inactivity at prednisolone 10 mg) in a similar proportion of patients when used as an initial agent or when switching to it after other agents (Table 3, available at www.aaojournal.org). The prior use of other
Table 5. Outcomes of Commonly Used Combination Agents Success (%)
VA Changes (%)
Reason for Discontinuation (%)
Agents Used
Prednisolone 10 mg
Prednisolone 7.5 mg
Gain 2 Lines
Loss 2 Lines
Ineffective
Side Effects
Remission
AZA þ CSA (n ¼ 14) CSA þ MMF (n ¼ 20) CSA þ MTX (n ¼ 4)
71 65 50
29 25 0
21 30 0
21 35 25
29 50 50
29 15 25
36 10 25
ALK ¼ alkylating agents; AZA ¼ azathioprine; CSA ¼ cyclosporine A; MMF ¼ mycophenolate mofetil; MTX ¼ methotrexate; VA ¼ visual acuity.
1122
Joshi et al
Outcomes of Changing Immunosuppression in Uveitis
Table 6. Outcomes of Biological Agents after Failure of Other Agents Success (%)
VA Changes (%)
Reason for Discontinuation (%)
Agent Used
Prednisolone 10 mg
Prednisolone 7.5 mg
Gain 2 Lines
Loss 2 Lines
Ineffective
Side Effects
Remission
Unknown
Infliximab (n ¼ 30) Adalimumab (n ¼ 8)
80 88
60 75
17 25
20 50
17 63
13 0
23 0
7 0
VA ¼ visual acuity.
ISAs in patients switching to CSA was not associated with differences in subsequent treatment success compared with those patients using CSA for the first time in the SITE cohort.5 The definition of success used in our study was similar to that used by the SITE cohort study. During their sensitivity analysis, the authors found that not having such a stringent definition of success (i.e., success to be sustained) also resulted in a larger proportion of patients achieving success on the basis of this definition. However, their analyses found that for some ISAs, 83% of patients who achieved control would subsequently have had a relapse by the next visit,2 suggesting that many patients relapse just after achieving control of inflammation or corticosteroid-sparing goals. Local and topical steroids were increased for many of the patients in our study, without systemic steroids being increased. Indications included an increase in anterior chamber cell activity (topical steroids) and the presence of cystoid macular edema (CME) and vitritis (local steroids). In the SITE cohort study, patients may have achieved quiescence if there was no cellular activity and prednisolone doses 10 mg, but there is no indication of whether these patients required an increase in local steroids to treat CME. If there were such patients, did they truly achieve “sustained control” of inflammation? Thus, such patients would actually overestimate the benefits of treatment effect. Whether CME is a measure of disease activity is a matter of debate. Cystoid macular edema certainly can be treated in a similar way as ocular cellular activity through the increase of oral corticosteroids,14-16 which may actually lead to a more rapid decrease of CME than with periocular steroid injections.17 However, the SUN group regards CME as a structural complication of uveitis,8 rather than a measure of active disease. The basis of this assumption shares common ground with the explanation for the presence of anterior chamber flare in chronic uveitis. The presence of CME and anterior chamber flare is due to vascular damage from the inflammatory process, and both may be long-standing, if not permanent; thus, both are believed to be a marker of past inflammation but not a sign of active inflammation.18,19
Study Limitations The limitations of our study include its retrospective nature, which can lead to imprecise and nonstandardized data collection from assessment at irregular intervals and differential follow-up. Treatment bias may have existed because treatment was not randomized or administered in a standard manner.
The SITE cohort study did not report VA outcomes, which can be difficult to report in a standardized manner. Several approaches can be taken. In studies in which the follow-up time is variable, the proportion with a particular threshold VA at presentation and rates of VA events during follow-up (e.g., falling below a specified threshold because of disease activity) can be reported,8,20 which is the method we adopted in our study. However, this can be problematic because a decline in VA may not be a direct result of the current ISA regimen that a patient is receiving. For example, a cataract may develop in a patient receiving a particular ISA, but although this ISA may have achieved sustained control for this patient and the decline may not be due to disease activity, any improvement in VA caused by the ISA regimen would be limited by the cataract.
References 1. Jabs DA, Rosenbaum JT, Foster CS, et al. Guidelines for the use of immunosuppressive drugs in patients with ocular inflammatory disorders: recommendations of an expert panel. Am J Ophthalmol 2000;130:492–513. 2. Gangaputra S, Newcomb CW, Liesegang TL, et al. Systemic Immunosuppressive Therapy for Eye Diseases Cohort Study. Methotrexate for ocular inflammatory diseases. Ophthalmology 2009;116:2188–98. 3. Pasadhika S, Kempen JH, Newcomb CW, et al. Azathioprine for ocular inflammatory diseases. Am J Ophthalmol 2009;148: 500–9. 4. Daniel E, Thorne JE, Newcomb CW, et al. Mycophenolate mofetil for ocular inflammation. Am J Ophthalmol 2010;149:423–32. 5. Kacmaz RO, Kempen JH, Newcomb C, et al. Cyclosporine for ocular inflammatory diseases. Ophthalmology 2010;117:576–84. 6. Pujari SS, Kempen JH, Newcomb CW, et al. Cyclophosphamide for ocular inflammatory diseases. Ophthalmology 2010;117:356–65. 7. Sobrin L, Christen W, Foster CS. Mycophenolate mofetil after methotrexate failure or intolerance in the treatment of scleritis and uveitis. Ophthalmology 2008;115:1416–21. 8. Standardization of Uveitis Nomenclature (SUN) Working Group. Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop. Am J Ophthalmol 2005;140:509–16. 9. Gregori NZ, Feuer W, Rosenfeld PJ. Novel method for analyzing Snellen visual acuity measurements. Retina 2010;30:1046–50. 10. Menezo V, Lau C, Comer M, Lightman S. Clinical outcome of chronic immunosuppression in patients with non-infectious uveitis. Clin Experiment Ophthalmol 2005;33:16–21.
1123
Ophthalmology Volume 121, Number 5, May 2014 11. Pascalis L, Pia G, Aresu G, et al. Combined cyclosporin Asteroid-MTX treatment in endogenous non-infectious uveitis. J Autoimmun 1993;6:467–80. 12. Kiss S, Ahmed M, Letko E, Foster CS. Long-term follow-up of patients with birdshot retinochoroidopathy treated with corticosteroid-sparing systemic immunomodulatory therapy. Ophthalmology 2005;112:1066–71. 13. Rothova A, Berendschot TT, Probst K, et al. Birdshot chorioretinopathy: long-term manifestations and visual prognosis. Ophthalmology 2004;111:954–9. 14. Rothova A. Inflammatory cystoid macular edema. Curr Opin Ophthalmol 2007;18:487–92. 15. McCluskey PJ, Towler HM, Lightman S. Management of chronic uveitis. BMJ 2000;320:555–8.
16. Okhravi N, Lightman S. Cystoid macular edema in uveitis. Ocul Immunol Inflamm 2003;11:29–38. 17. Venkatesh P, Abhas Z, Garg S, Vohra R. Prospective optical coherence tomographic evaluation of the efficacy of oral and posterior subtenon corticosteroids in patients with intermediate uveitis. Graefes Arch Clin Exp Ophthalmol 2007;245:59–67. 18. Samson CM, Waheed N, Baltatzis S, Foster CS. Methotrexate therapy for chronic noninfectious uveitis: analysis of a case series of 160 patients. Ophthalmology 2001;108:1134–9. 19. Chorich LJ III, Klisovic DD, Foster CS. Diagnosis of uveitis. In: Foster CS, Vitale AT. Diagnosis and Treatment of Uveitis. 2nd ed. New Delhi: Jaypee Brothers Medical; 2013:101–30. 20. Jabs DA. Improving the reporting of clinical case series. Am J Ophthalmol 2005;139:900–5.
Footnotes and Financial Disclosures Originally received: June 11, 2013. Final revision: November 16, 2013. Accepted: November 18, 2013. Available online: January 14, 2014.
Manuscript no. 2013-933.
1
Moorfields Eye Hospital, London, United Kingdom.
2
UCL Institute of Ophthalmology, London, United Kingdom.
3
Royal Surrey County Hospital, Guildford, United Kingdom.
4
Save Sight Institute, Sydney Eye Hospital, Sydney, Australia.
5
Faculty of Medicine, Imperial College London, Hammersmith Hospital, London, United Kingdom.
6
School of Medical Sciences, University of New South Wales, Kensington, New South Wales, Australia.
1124
Financial Disclosure(s): The author(s) have made the following disclosure(s): S.R.J.T. has board membership with Allergan and has received financial support for the development of educational presentations by Allergan. S.L. has board membership with Allergan and GlaxoSmithKline. Funding: L.J. and S.R.J.T. were supported by the UK National Institute of Health Research. S.L. has received consultancy fees from PAREXEL International Corp., Bayer, and Zeiss, and received payment for lectures and development of educational materials from Allergan. Correspondence: Susan Lightman, PhD, FRCOphth, UCL Institute of Ophthalmology, 11-43 Bath Street, London, EC1V 9EL. E-mail: [email protected].
