The annual relapse rate reduced after MMF treatment in 51 of 58 patients with neuromyelitis optica spectrum disorder during a median follow-up time of 20 months.
eTable. Summarization of the outcome of previous treatments, MMF, and rituximab treatment
Huh S, Kim S, Hyun J, Joung A, Park MS, Kim B, Kim HJ. Mycophenolate Mofetil in the Treatment of Neuromyelitis Optica Spectrum Disorder. JAMA Neurol. 2014;71(11):1372-1378. doi:10.1001/jamaneurol.2014.2057
Neuromyelitis optica spectrum disorder (NMOSD) is an autoimmune inflammatory disorder of the central nervous system. Recently, various immunosuppressant medications were introduced as therapeutic options for preventing relapse of NMOSD. However, our understanding of the effectiveness of mycophenolate mofetil (MMF) in treating patients with NMOSD is based on only a small number of studies.
To evaluate the efficacy and safety of MMF treatment in patients with NMOSD.
Design, Setting, and Participants
A 3-center retrospective review of our experiences, examining results from 59 patients with NMOSD (24 with neuromyelitis optica and 35 with a limited form of the disease) who were treated with MMF (1000-2000 mg/d).
Main Outcomes and Measures
Patients’ annualized relapse rate, disability as measured by the Expanded Disability Status Scale, and experience of adverse effects due to MMF were assessed.
Of the 59 patients, 1 with NMOSD who had to discontinue MMF use in the first month due to a rash was excluded. The remaining 58 patients were included in the drug-efficacy analysis. The median post-MMF annualized relapse rate was significantly lower than the pre-MMF annualized relapse rate (0.0 vs 1.5; P < .001). The Expanded Disability Status Scale scores also significantly decreased after MMF treatment (3.0 vs 2.5; P = .005). Thirty-five patients (60%) were relapse free, and Expanded Disability Status Scale scores were stabilized or improved in 53 patients (91%). Fourteen patients discontinued MMF treatment owing to ongoing relapse (10 patients), rash (1 patient), pregnancy (1 patient), and financial problems (2 patients), but MMF was generally well tolerated.
Conclusions and Relevance
In this observational study, MMF treatment induced reduction of relapse frequency, stabilized or improved disability, and was well tolerated in patients with NMOSD.
Quiz Ref IDNeuromyelitis optica (NMO) spectrum disorder (NMOSD) is an autoimmune inflammatory disorder of the central nervous system that is characterized by severe optic neuritis or longitudinal extensive transverse myelitis. The disease course of NMOSD is often more devastating than that of multiple sclerosis. Unlike multiple sclerosis, disabilities due to NMOSD arise from acute attacks, and just 1 or 2 attacks can lead to blindness or an ambulatory disability.1,2 Therefore, prevention of attacks is the key to avoiding attack-related disability. Quiz Ref IDSince the use of azathioprine (AZA) with prednisone was first reported in 1998,3 various immunosuppressants, including mycophenolate mofetil (MMF),4,5 methotrexate,6 rituximab,7- 15 and mitoxantrone hydrochloride,16,17 have been introduced as therapeutic options for preventing relapse in patients with NMOSD. Azathioprine is a commonly used immunosuppressant for patients with NMOSD, in which treatment has resulted in a reduction of the annual relapse rate (ARR).3,5,18,19 However, 53% of patients who received AZA reported new inflammatory central nervous system events,5 31% to 39% showed deterioration on the Expanded Disability Status Scale (EDSS),18,19 and approximately 38% had to discontinue treatment owing to adverse effects or insufficient functional effects.18 Rituximab, which recently has been regarded as a promising treatment for NMOSD, has limited availability owing to its considerable cost and the need to conduct surveillance of B cells by flow cytometry for tailored therapy. Mitoxantrone should be used as rescue treatment for patients with aggressive and refractory courses because of its dose-limiting adverse effects.20- 22 Therefore, there is an unmet need for alternative agents with more balanced safety, potency, and accessibility.
Mycophenolate mofetil, a prodrug of mycophenolic acid, has been used to treat psoriasis since 1970 and was approved by the US Food and Drug Administration for the prevention of renal transplant rejection. The use of MMF in treating NMOSD is based on only a few studies that showed some benefit in reducing relapse rates.4,5 The first was a retrospective, single-center, 24–case series that included only 7 treatment-naive patients; 33% of the included patients were receiving prednisone or intravenous immunoglobulin as concomitant therapy.4 The latest study, which included 28 patients who were prescribed MMF as initial treatment, mainly focused on comparing changes in ARR among patients with NMO who received AZA, MMF, and rituximab.5 In this study, we aimed to expand the field of knowledge by evaluating the efficacy and safety profile of MMF as an initial or alternative therapy in a larger cohort of patients with NMO and NMOSD from 3 referral centers in Korea.
We retrospectively reviewed the medical records of patients with NMO or NMOSD (2006 diagnostic criteria23 or NMOSD24) who were treated with MMF in 3 referral centers in Korea from March 1, 2009, through March 31, 2013. The patients were identified as having been treated with MMF for at least 6 months. However, even with less than 6 months’ duration of treatment, the patients who discontinued taking MMF owing to adverse effects or ongoing relapses are described and included to assess tolerability. This study was approved by the institutional review board or independent ethics committee at the National Cancer Center, Yeungnam University College of Medicine, and Korea University Medical Center.
The clinical course, number of relapses, administered drug information before the initiation of MMF treatment, adverse effects, EDSS score before and after the latest MMF treatment, concomitant treatments, and dosage of MMF were compiled by medical record review. The efficacy was estimated by ARR and EDSS score changes between the pre-MMF and post-MMF treatment periods. A relapse was defined as a neurological disturbance that increased the EDSS score by at least half a point or when worsening of 1 point in 2 functional systems or 2 points in 1 functional system occurred and lasted for at least 24 hours in the absence of fever or infection.
The rescue therapies during acute relapse, such as plasmapheresis or intravenous methylprednisolone with prednisolone tapering within 2 months, were excluded from MMF concomitant treatment. We also subanalyzed MMF efficacy depending on the prior use of immunosuppressants for at least 1 month.
Data were analyzed using GraphPad Prism, version 4.0 (GraphPad Software, Inc). The paired t test or Wilcoxon signed rank test was used for the comparison of ARR and EDSS scores before and after MMF treatment. P < .05 was considered statistically significant.
Of the 59 patients identified, 1 patient with NMOSD who had to discontinue MMF in the first month owing to a rash was excluded. The remaining 58 patients were included in the efficacy analysis. The demographic and clinical characteristics of the cohort are summarized in Table 1. Forty-four patients were treated with 2000 mg/d of oral MMF, 13 with 1500 mg/d, and 1 with 1000 mg/d. One patient was treated with 10 mg/d of concomitant prednisolone to maintain treatment for 4 months.
The median number of attacks before the initiation of MMF treatment was 3 (range, 1-23) for a median duration of 36 months. While taking MMF, the median number of attacks was 0 (range, 0-6) for a median duration of treatment of 20 months. The median pretreatment ARR was 1.5, and the median posttreatment ARR was 0.0 (P < .001) (Table 2). Treatment with MMF resulted in a reduction of ARR in 51 of 58 patients (88%). The median posttreatment EDSS scores were improved compared with the median pretreatment EDSS scores. The EDSS scores were improved or stabilized in 53 of 58 of patients (91%; Table 2).
Thirty-six patients were treated with MMF as an initial immunosuppressive agent. The median ARR decreased from 2.2 to 0.0 (P < .001), and the median EDSS score decreased from 3.0 to 2.3 (P = .02) for a median duration of treatment of 23 months (Table 2). A total of 32 patients (89%) showed a reduction of ARR, and 32 patients improved or maintained their EDSS score; 24 patients (67%) were relapse free, while 17 attacks occurred in 12 patients during MMF treatment.
Twenty-two patients (38%) had received 1 or more immunosuppressants for at least 1 month before starting MMF treatment, as follows: AZA alone or with prednisolone (n = 12), mitoxantrone (n = 8), fingolimod (n = 1), and hydroxychloroquine sulfate (n = 1). In these patients, the median ARR decreased from 1.2 to 0.1 (P < .001) for a median duration of treatment of 19 months. The change of the EDSS score was not significant (P = .07) (Table 2). Of the 22 patients, 11 (50%) were relapse free, while 18 relapses occurred in 11 patients during MMF treatment.
In this study, 12 patients were treated with AZA before the initiation of MMF treatment. In total, during the entire period, these patients showed an improvement in relapse rate (median ARR before MMF vs after, 1.1 vs 0.0; P = .02), and 7 were relapse free during MMF treatment (Table 2). Of the 12 patients, when we subanalyzed details, such as number of attacks and treatment period, according to the type of medication, the median ARR was significantly reduced after MMF treatment compared with the period before receiving immunosuppressants and AZA treatment (ARR before treatment vs after receiving MMF, 1.5 vs 0.0; P = .02; AZA vs MMF, 0.6 vs 0.0; P = .02). With a median duration of AZA treatment of 14 months, the median ARR was not significantly reduced compared with the period before immunosuppressant use (P = .27). Eleven relapses occurred in 9 patients despite AZA treatment; treatment with AZA was changed to MMF because of relapses (n = 8), adverse effects of AZA (n = 2), adverse effects of concomitant prednisolone (n = 1), and consistently high NMO-antibody titer (n = 1). While receiving MMF treatment, the median ARR decreased to 0, and 7 relapses occurred in 5 patients (42%). Of these, 2 patients switched to rituximab treatment at the last follow-up owing to ongoing relapses.
Eight patients received MMF as maintenance therapy following mitoxantrone treatment. Although there was a reduction in the median ARR after MMF treatment (before vs after ARR, 1.7 vs 0.4; P = .02), only 2 patients were relapse free and 6 had relapses. Of these, 3 patients switched to rituximab treatment owing to ongoing relapses. There was no interval change in the median EDSS score after MMF treatment (EDSS score before vs after MMF treatment, 4.5 vs 4.5; P = .07) for a median duration of MMF treatment of 29 months (Table 2), and no patient had further worsening of EDSS scores.
Thirty-five patients (60%) were relapse free, with a median MMF treatment duration of 20 months (Figure). Of these, 1 patient stopped taking MMF at 25 months to prepare for pregnancy; she had a relapse after 4 months of discontinuation and resumed MMF treatment.
Twenty-three patients (40%) experienced 35 relapses while taking MMF. Of these, 11 patients reported inadequate MMF medication effects (6 patients did not take medication consistently, and 5 experienced relapses during the first month after the initiation of MMF treatment with a low dosage). The remaining 12 patients had relapses despite adequate MMF treatment. Ten of 23 patients with relapse were switched to rituximab treatment at their last follow-up. The remaining 13 patients were continuously treated with MMF because they had minor relapses, and they recovered to their former functional levels after intravenous methylprednisolone pulse therapy.
Treatment with MMF was generally well tolerated, and only 1 case of rash led to discontinuation within 1 month of treatment. Of the 59 patients, 14 experienced adverse events associated with MMF including rash (n = 1), amenorrhea (n = 2), herpes zoster (n = 1), cystitis (n = 3), pneumonia (n = 1), hypotension (n = 1), fatigue (n = 1), and mild hair loss (n = 4).
The purpose of this study was to assess the efficacy and tolerability of MMF treatment in patients with NMOSD. Quiz Ref IDWe demonstrated that the median ARR was significantly reduced in 51 0f 58 patients (88%), and disability was improved or maintained in 53 patients (91%). In addition, 60% of patients were relapse free after MMF treatment, and 78% were continuing MMF treatment at the last follow-up. Of the 23 patients who had relapses during MMF treatment, 11 of the relapses were attributed to inadequate immunosuppressive effects. When we excluded these 11 patients, the relapse-free rate increased to 74% (Table 2).
In a study by Jacob et al,4 MMF treatment in 24 patients with NMOSD with a median treatment duration of 27 months resulted in a significant reduction in ARR (ARR before vs after, 1.28 vs 0.09; P < .001); stabilization or decrease in EDSS score in 91%; and adverse effects during treatment with MMF in 25%. These results are similar to ours, with the exceptions of the higher median EDSS score (median EDSS score, 6 vs 3), higher rate of patients who received MMF with concomitant therapy (33% vs 2%), and a lower percentage of treatment-naive patients (29% vs 52%) in their study vs ours, respectively (Table 3). The latest study,5 which involved 28 patients with NMO who took MMF as an initial immunosuppressive agent, showed that 64% were relapse free and 87.4% showed ARR reduction. These results are consistent with those of our study, in which 36 patients were immunosuppressant naive. Of these, 24 (67%) were relapse free and 32 (89%) showed ARR reduction.
Although it is difficult to make a direct comparison between MMF and other immunosuppressive agents in terms of maintenance therapies for NMOSD, recent AZA efficacy studies18,19 showed a reduction of the median ARR, whereas the relapse-free rate and rate of stabilized or improved EDSS scores were relatively lower than observed in our study (relapse-free rate, 37%-55% vs 60%; stabilized or improved EDSS score, 61%-69% vs 91%) (Table 3).
Quiz Ref IDIn our study, 12 patients who had been treated with AZA before MMF showed a significant reduction in ARR after switching to MMF treatment (Table 2). Although this analysis was performed in a small number of patients, these observations suggest that MMF can be a good option in patients who experienced relapse or adverse effects during AZA treatment. In addition, the discontinuance rates of AZA were reported as 38% and 46% in the studies by Costanzi et al18 and Elsone et al,25 respectively, higher than the MMF discontinuance rate in our study (24%). Mycophenolate mofetil selectively inhibits the growth of T and B cells and does not affect other immune cells because other cells are able to recover purines via a separate salvage pathway.26 In addition, Eickenberg et al27 investigated patients with systemic lupus erythematosus and reported that transitional and naive B-cell counts were significantly higher in patients who were taking MMF compared with those who were taking AZA, but markedly lower frequencies of plasmablasts were detected in patients who were taking MMF compared with patients who were taking AZA or were untreated. Therefore, the sparing of immune cells other than lymphocytes and relative sparing of antigen-naive B-cell subsets may contribute to favorable safety and tolerability of MMF compared with AZA.
When compared with outcomes in rituximab7,8,14 and mitoxantrone16 efficacy studies, our MMF-treated patients had similar ARR after treatment and relapse-free rates but tended to have lower ARR and EDSS scores before treatment than in other immunosuppressant studies (Table 3). Our study comprised only 7 patients (12%) with severe neurological impairments (EDSS score >6), whereas the studies that used mitoxantrone16 and rituximab11 consisted of 26% to 45% of those cases. Therefore, it is difficult to ascertain whether the treatment with MMF has enough effectiveness in advanced NMOSD. For this reason, we subanalyzed a group of 8 patients with NMOSD with a high relapse rate who used MMF as maintenance therapy after mitoxantrone. Interestingly, only 2 of these 8 patients (25%) were relapse free, and the remaining 6 had relapses. Of these, 3 finally switched from MMF treatment to rituximab owing to ongoing relapses. In total, 10 of 58 patients switched from MMF to rituximab treatment owing to ongoing relapse; of these, 6 patients showed a reduction of ARR and 3 were relapse free for a median duration of rituximab treatment of 23 months. A detailed description of the patients who had switched to rituximab treatment is available in the eTable in the Supplement. These observations suggest that MMF may not be potent enough to inhibit relapses in patients with high disease activity. Therefore, other treatment options should be considered when major relapses occur repeatedly in patients treated adequately with MMF.
Five patients (8%) experienced relapses during the first month after the initiation of MMF treatment with less optimal dosage. All patients received prednisolone tapering (10-60 mg) for 2 months after receiving intravenous methylprednisolone for acute relapse. Those patients were continuously treated with MMF when disability was completely recovered to the former level after the second intravenous methylprednisolone therapy. Therefore, to prevent such early relapses, a relatively high dosage of concomitant prednisolone seems necessary for at least a few months after initiation of MMF treatment.
Various adverse events, including diarrhea, infections, amenorrhea, alopecia, hypercholesterolemia, myelotoxicity, and gastrointestinal complications, have been reported in patients treated with MMF.28- 30 However, the use of MMF was associated with fewer infections and did not increase the risk of malignant neoplasms compared with other immunosuppressive agents.28,29 Consistent with those articles, our study also showed excellent tolerability in most patients. The discontinuation of MMF treatment owing to adverse effects (1 of 59 patients) was much less common than it was in an AZA efficacy study for NMOSD (25 of 99 patients).18 Moreover, 52% of patients received combination treatment with prednisolone in an AZA study for NMOSD,18 whereas 57 patients (98%) received MMF monotherapy in our study, which was likely to accompany good adherence.
Our study had some limitations. First, it was a retrospective study and lacked the randomization observed in clinical trials. Second, there was a potential unintentional selection bias because our study included patients with relatively lower disease activity because patients were recruited from referral centers that tend to use rituximab or mitoxantrone in advanced or highly relapsing cases.14,16 Third, the effectiveness of MMF could have been influenced by previous immunosuppressants or immunomodulators, and those residual effects could provide either beneficial or harmful effects.
Quiz Ref IDDespite these limitations, the results suggest that the use of MMF can result in good disease control with excellent safety and tolerability in patients with NMOSD. The next step will be to conduct a randomized clinical trial with long-term efficacy and safety analyses in a larger cohort that can elucidate the efficacy of MMF treatment in patients with NMOSD.
Accepted for Publication: June 11, 2014.
Corresponding Author: Ho Jin Kim, MD, PhD, Department of Neurology, Research Institute and Hospital of National Cancer Center, 323 Ilsan St, Ilsandong-gu, Goyang-si, Gyeonggi-do, Korea (email@example.com).
Published Online: September 8, 2014. doi:10.1001/jamaneurol.2014.2057.
Author Contributions: Drs H. J. Kim and Huh had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Huh, H. J. Kim.
Acquisition, analysis, or interpretation of data: Huh, S.-H. Kim, Hyun, Joung, Park, B.-J. Kim, H. J. Kim.
Drafting of the manuscript: Huh, S.-H. Kim, Park, H. J. Kim.
Critical revision of the manuscript for important intellectual content: Huh, S.-H. Kim, Hyun, Joung, B.-J. Kim, H. J. Kim.
Statistical analysis: Huh, H. J. Kim.
Administrative, technical, and material support: Hyun, Joung.
Study supervision: S.-H. Kim, B.-J. Kim, H. S. Kim.
Conflict of Interest Disclosures: Dr H. J. Kim reported having given talks, consulted, and received honoraria and/or hospitality and/or support for research activities from Bayer Schering Pharma, Biogen Idec, Genzyme, GemVax and KAEL, MedImmune, Merck Serono, Novartis, Teva-Handok, and Union Chimique Belge. No other disclosures were reported.