Study identification and selection
There were 1167 articles identified through database searching from Ovid MEDLINE and EMBASE. After excluding 170 duplicates, a total of 997 studies were screened by titles and abstracts. Forty-two studies were analyzed for eligibility assessment. Of these 42 studies, 27 studies were excluded (2 studies were reviews, 5 studies were case reports or case series with fewer than three patients, 7 studies had no reported ARR or EDSS pre- and post- treatment, 4 studies were duplicates, 8 studies had no full-text available, and 1 study had no English article available). As a result, 15 studies (10 retrospective and 5 prospective) published from 2009–2020 met our study criteria and were included in the systemic analysis. (Fig. 1) None were randomized controlled trials, 14 were cohort studies, and 1 was a case series
Assessment of risk of bias
Demographic and clinical characteristics
The characteristics and demographic data for the 15 studies are described in Table 1 and the supplementary table. There were a total of 1047 patients (799 of them were treated with MMF). The total number of female patients was 915 (overall 87.4%) with the female proportion varying from 73.8–93.3% for each individual study. The AQP4-IgG serostatus was reported for all patients, of whom 886 patients (84.6%) were AQP4-IgG seropositive. Diagnosis of NMO/NMOSD was given according to the 2006 [2], 2007 [11], or 2015 [1] criteria.
Table 1
Baseline characteristics of 15 studies in neuromyelitis optica spectrum disorders treated with mycophenolate mofetil
Author | Study design | Diagnosis of NMO/ NMOSD | Number with positive AQP4 antibody/Total number (%) | Number of females/Total number (%) | Number of patients treated with MMF | Age of onset, years old | Dose of MMF | Other immune-suppressive (IS) therapy prior to MMF; number of patients (%) | Concurrent use of corticosteroid; number of patients (%) | Assessment of treatment response and duration of follow-up/treatment |
Jacob et al. 2009 [8] | Retrospective case series | The 2006 NMO criteria OR The 2007 NMOSD criteria | 23/24 (95.8%) | 19/24 (79.2%) | 24 | Median 56 (range 34–77) | Median 2000 mg/day (range 750–3000) | 17 (70.8%) | 9 (37.5%) | Assessment: at final follow-up visit Duration of follow-up: median 28 months (range 18–89 months) |
Huh et al. 2014 [24] | Retrospective cohort | The 2006 NMO criteria OR The 2007 NMOSD criteria | 52/58 (89.7%) | 50/58 (86.2%) | 58 | Median 34 (range 10–53) | 1000–2000 mg/day | 22 (37.9%) | 1 (1.72%) | Assessment: at latest MMF treatment Duration of treatment: median 20 months (range 4–67 months) |
Mealy et al. 2014 [26] | Retrospective cohort | The 2006 NMO criteria OR The 2007 NMOSD criteria | 17/28 (60.7%) | 26/28 (92.9%) | 28 | Median 36 (range 19–74) | 1000–2000 mg/day | 8 (28.6%) | 13 (46.4%) | Duration of treatment: median 26 months (range 6–68 months) |
Torres et al. 2015 [20] | Retrospective cohort | The 2006 NMO criteria OR The 2007 NMOSD criteria | 4/11 (36.4%) | 10/11 (90.9%) | 11 | Median 37 (range 18–68) | NR | 7 (63.6%) | NR | Duration of follow-up: median 23 months (range 13–60 months) |
Chen et al. 2016 [13] | Prospective cohort | The 2006 NMO criteria OR The 2007 NMOSD criteria | 52/62 (83.9%) | 58/62 (93.5%) | 62 | Mean 38.7 (SD 12.0) | 20 mg/kg | 7 (11.3%) | 24 (38.7%) | Assessment: at final follow-up visit Duration of follow-up: median 18 months (range 6–49 months) |
Jeong et al. 2016 [15] | Retrospective cohort | The 2006 NMO criteria OR The 2007 NMOSD criteria | 32/34 (94.1%) | 29/34 (85.3%) | 34 | Median 35 (range 10–53) | 1500–2000 mg/day | None (0%) | 9 (26.4%) | Duration of treatment: median 26.1 months (range 5.5–68.6 months) |
Xu et al. 2016 [16] | Prospective cohort | The 2015 IPND | 33/38 (86.8%) | 32/38 (84.2%) | 38 | Mean 28.7 (SD 13.0) | 1500 mg/day | None (0%) | All (100%) | Duration of treatment: median 15.2 months (range 6.6–26.4 months) |
Chen et al. 2017 [14] | Prospective cohort | The 2006 NMO criteria OR The 2007 NMOSD criteria | 89/105 (84.8%) | 97/105 (92.4%) | 105 | Mean 44.0 (SD 12.1) | 20 mg/kg/d | None (0%) | 49 (46.6%) | Assessment: at final follow-up visit Duration of treatment: median 17 months (range 6–78 months) |
Montcuquet,et al. 2017 [17] | Retrospective Cohort | The 2015 IPND | 45/67 (67.2%) | 50/67 (74.6%) | 67 | Median 37.9 (range 6–67) | 2000 mg/day | None (0%) | 16 (23.9%) | Duration of treatment: median 24 months (range 1-156 months) |
Huang et al. 2018 [22] | Prospective cohort | The 2006 NMO criteria OR The 2015 IPND | 90/90 (100%) | 84/90 (93.3%) | 90 | Median 36 (range 10–65) | 1000 mg/day | 20 (22.2%) | All (100%) | Duration of follow-up: median 13.5 months |
Jiao et al. 2018 [12] | Retrospective cohort | The 2006 NMO criteria OR The 2007 NMOSD criteria | 74/86 (86.0%) | 77/86 (89.5%) | 86 | Median 43 (range 6–68) | High dose (1750–2000 mg) Moderate dose (1250–1500 mg) Low dose (≤ 1000 mg) | 56 (65.1%) | 65 (76%) | Assessment: at final follow-up visit Duration of treatment: median 20 months (range 6–89 months) |
Mealy et al. 2018 [21] | Retrospective cohort | The 2015 IPND | 208/245 (84.9%) | 216/245 (88.2%) | 103 | Median 37 (range 7–79) | 1500–2000 mg/day | Some had glatiramer acetate | None (0%) | Duration of treatment: median 36 months (range 6–92 months) Duration of follow-up: median 95 months (mean, 103 months) |
Yang et al. 2018 [18] | Prospective cohort | The 2015 IPND | 13/30 (43.3%) | 26/30 (86.7%) | 30 | Mean 42.6 (SD 11.7) | 1000 mg/day | None (0%) | 28 (93.3%) | Assessment: at final follow-up visit Duration of follow-up: median 28.5 months (range 19–42 months) |
Zhou et al. 2019 [23] | Retrospective cohort | The 2006 NMO criteria OR The 2015 IPND | Pediatric group: 23/31 (74.2%) | Pediatric group: 25/31 (80.6%) | 4 | Pediatric group: Median 14 (range 10–17) | 1000 mg/day | Some had AZA or CYP | All (100%) | Assessment: at final follow-up visit Duration of follow-up: median 17 months (range 8–26 months) |
Adult group: 96/96 (100%) | Adult group: 85/96 (88.8%) | 17 | Adult group: Median 35 (range 18–96) | 1000 mg/day | All (100%) |
Poupart et al. 2020 [19] | Retrospective cohort | The 2015 IPND | 35/42 (83.3%) | 31/42 (73.8%) | 42 | Mean 41.4 (SD 17.6) | 1000–2000 mg/day | None (0%) | 8 (19.1%) | Median 35 months (interquartile range 3.2) |
Abbreviations: AQP4: Aquaporin4; AZA: azathioprine; CYP: cyclophosphamide; IS: immunosuppressive; kg: kilogram; mg: milligram; MMF: mycophenolate mofetil; NMO: neuromyelitis optica; NMOSD: neuromyelitis optica spectrum disorders; NR: not reported; ON: optic neuritis; SD: standard deviation; IPND: International Panel for Neuromyelitis optica Diagnosis |
Treatment regimens
MMF was administered 1000–2000 mg/day in most of the studies. One study by Jiao et al., reported categorized dosages (1000 mg/day or less, 1250–1500 mg/day, 1750–2000 mg/day as low, moderate and high dose respectively). [12] Studies by Chen et al. prescribed 20 mg/kg dosage. [13, 14] Of the 799 patients treated with MMF, MMF was used as a first-line therapy in 6 studies [15, 16, 14, 17–19] with a total of 316 patients (39.5%). The other 9 studies included patients who had suboptimal treatment from prior IS including AZA, CYP, mitoxantrone, fingolimod. hydroxychloroquine sulfate, beta-interferons, and glatiramer acetate, however contained no detailed information of the proportionate use. Data on concomitant corticosteroids were available in all but one study [20]. The proportion of steroid use ranged from 0% [21] to 100% [16, 22, 23]. In four studies [16, 22, 23, 18], 177 patients (22.2%) had been taking oral corticosteroid at the time of MMF treatment. Two studies [21, 24] were conducted with 160 patients (20.0%) who did not receive corticosteroid during MMF treatment. The remaining 10 studies contained both groups of patients.
Treatment outcome analysis
The efficacy of MMF treatment determined by changes in ARR and EDSS is shown in Table 2. The median follow-up duration ranged from 13.5 months [22] to 95 months [21], with less than 24 months in 8 studies and 24 months or more in 7 studies. (Table 1)
Table 2
Changes in Expanded Disabilities Status Score and annualized relapse rate after treatment with mycophenolate mofetil
Author | EDSS | Improved or stabilized EDSS (%) | ARR | Relapse free rate (%) |
Median Pre-treatment (Range) | Median Post-treatment (Range) | P-value | Median Pre-treatment (Range) | Median Post-treatment (Range) | P-value |
Jacob et al. 2009 [8] | 6.0 (0.0–8.0) | 5.5 (0.0–10) | 0.17 | 91% | 1.15 (0.23–11.78) | 0.18 (0.00-1.50) | < 0.01 | 46% |
Huh et al. 2014 [24] | 3.0 (0.0–8.0) 3.2 (2.2)a | 2.5 (0.0–7.0) 2.7 (1.9)a | 0.01 | 91% | 1.50 (0.30–11.80) 2.6 (2.7)a | 0.00 (0.00-2.60) 0.5 (0.8)a | < 0.01 | 60% |
Mealy et al. 2014 [26] | NR | NR | NR | NR | 2.61 (NR) | 0.33 (NR) | < 0.01 | 64% |
Torres et al. 2015 [20] | 4.0 (3.0-6.5) | 5.0 (NR) | 0.46 | NR | 1.06 (0.84–2.31) | 0.39 (NR) | < 0.05 | 27% |
Chen et al. 2016 [13] | 4.0 (0.5-8.0) 4.1 (2)a | 2.0 (0.5–7.5) 2.8 (2.1)a | < 0.01 | 95.2% | 1.20 (0.20-7.00) 1.7 (1.2)a | 0.00 (0.00-1.70) 0.4 (0.5)a | < 0.01 | 58.1% |
Jeong et al. 2016 [15] | 3.0 (0.0–7.0) | 2.0 (0.0–7.0) | < 0.01 | NR | 1.54 (NR) | 0.18 (NR) | < 0.01 | 64.7% |
Xu et al. 2016 [16] | 2.0 (0.0–9.0) 2.7 (2)a | 2.0 (0.0-8.5) 2.0 (1.8)a | < 0.01 | 97.4% | 0.80 (0.00–8.00) 1.0 (1.0)a | 0.00 (0.00-1.40) 0.1 (0.3)a | 0.05 | NR |
Chen et al. 2017 [14] | 3.0 (0.5-8.0) | 2.0 (0.5–7.5) | < 0.01 | NR | 1.20 (0.10-7.00) | 0.00 (0.00–2.00) | < 0.01 | 56.2% |
Montcuquet et al. 2017 [17] | 4.0 (0.0-8.5) | 3.8 (0.0–10.0) | < 0.05 | NR | 1.00 (0.10–3.20) | 0.00 (0.00–3.00) | < 0.05 | 49.3 |
Huang et al. 2018 [22] | 4.0 (0.0-8.5) | 3.0 (0.0-8.5) | < 0.01 | 90% | 1.02 (0.00-19.21) | 0.00 (0.00-2.44) | < 0.01 | 73% |
Jiao et al. 2018 [12] | 3.0 (0.0-8.5) | 2.5 (0.0-8.5) | 0.01 | 87% | 1.40 (0.10–11.00) | 0.00 (0.00-2.80) | < 0.01 | 64% |
Mealy et al. 2018 [21] | NR | NR | NR | NR | 1.79 | 0.29 | < 0.01 | 64.7% |
Yang et al. 2018 [18] | 3.5 (2.0-8.5) | 2.0 (0.5-7.0) | < 0.01 | 100% | 0.90 (0.00–5.00) | 0.00 (0.00-2.40) | < 0.01 | 60% |
Zhou et al. 2019 [23] | NR | NR | NR | NR | 1.00 (0.23–3.43) in adult patients | 0.00 (0.00–0.71) in adult patients | < 0.01 | 80% in adult patients |
0.98 (0.35–2.11) in pediatric patients | 0.28 (0-0.71) in pediatric patients | 50% in pediatric patients |
Poupart et al. 2020 [19] | NR | NR | NR | NR | 0.71 (0.43–1.15)b | 0.20 (0.11–0.35)b | NR | NR |
amean (SD) bmean (95% CI) |
Abbreviations: EDSS, Expanded Disability Status Scale; ARR, Annual Relapse Rate; NR, not reported |
Only 4 studies [24, 16, 13, 19] reporting ARR and 3 studies [24, 16, 13] reporting EDSS reported values by mean and SD, and therefore were included in the meta-analysis. (Table 2)
Qualitative analysis
All 15 studies reported median ARR before and after treatment. All but one [19] demonstrated significant ARR reduction after receiving MMF (p < 0.05). The relapse-free rate was 60% (ranged from 27–80%).
For the 11 studies reporting EDSS as a treatment outcome, 7 studies revealed stabilization or improvement of disability in patients receiving MMF treatment measured by EDSS with the proportion varying from 87–100%. 11 studies reported median EDSS before and after MMF treatment. Of those, 9 showed significant post-treatment median EDSS changes (p < 0.05).
Meta-analysis: Efficacy on the reduction of ARR
All the 4 studies [24, 16, 13, 19] including 200 NMOSD patients with the majority of patients being AQP4-positive (80–90%) showed a significant ARR reduction with the mean reduction of 1.13 (95% CI, 0.60 to 1.65) after MMF therapy with a dosage between 1000–2000 mg/day for 15.2–35 months, compared to the ARR at initiation of treatment. (Fig. 3a)
Meta-analysis: Efficacy on the EDSS
All the 3 studies [24, 16, 13, 19] with 158 NMOSD patients showed a significant EDSS lowering after MMF therapy with a mean reduction of 0.85 (95% CI, 0.36 to 1.34). Moreover, the Chen study [13] showed a large decrease in disability measured by EDSS from baseline of moderately severe (EDSS 4.1) to full independent (EDSS 2.1). (Fig. 3b)
Safety
ADRs are summarized in Table 3. For all 799 patients, data on safety of MMF therapy were recorded for 594 patients. 106 (17.8%) patients were reported to have developed ADRs. One of the most common ADRs were infections (33 patients; 5.6%)—including respiratory infection/pneumonia (12 patients; 2.0%), urinary tract infection (8 patients; 1.3%), herpes zoster infection (8 patients; 1.3%), herpes simplex infection (2 patients; 0.3%), and abnormal liver function tests (27 patients; 4.5%). The other common ADRs were hair loss (17 patients; 2.9%), gastrointestinal (GI) side effects (14 patients; 2.4%)—including nausea, diarrhea /abdominal pain, and constipation—bone marrow suppression (16 patients; 2.7%)—including anemia (6 patients; 1.0%), agranulocytosis (1 patient; 0.2%), leukopenia (8 patients; 1.3%), thrombocytopenia (1 patient; 0.2%), and amenorrhea in 3 patients (0.5%). Uncommon documented side effects were headaches, phlegm on normal CT chest, chronic dermopathy of the hand, rectal cancer, renal insufficiency, rash, hypotension, fatigue, easy bruising, anxiety, and sun sensitivity. The data on discontinuation of MMF were available for 687 patients, 27 (3.9%) of them discontinued MMF from ADRs such as rash, agranulocytosis, leukopenia, thrombocytopenia, arthromyalgia, GI side effects, and amenorrhea. All ADRs were reversible after discontinuation of MMF. One was discovered with high CERA. Three patients died during MMF treatment: one with EDSS 8.5 succumbed from complications of immobilization, another developed disseminated varicella zoster with acute respiratory distress syndrome, and the other had—according to death certificate documents—“cardiopulmonary failure; respiratory drive failure and Devic’s disease”. [17, 22, 8]
Table 3
Adverse events in 15 studies on mycophenolate mofetil in neuromyelitis optica spectrum disorders
Author | Total number of patients | Number of patients with adverse events (%) | Adverse events | Number of events (%) | Total number of discontinuation due to adverse effects (%) |
Jacob et al. 2009 [8] | 24 | 6 (25%) | Headache | 1 (4.2%) | 1 (4.2%) due to low white blood cell counts |
Constipation | 1 (4.2%) |
Easy bruising | 1 (4.2%) |
Anxiety | 1 (4.2%) |
Hair loss | 1 (4.2%) |
Diarrhea and abdominal pain | 1 (4.2%) |
Low white blood cell counts | 1 (4.2%) |
Huh et al. 2014 [24] | 58 | 14 (24.13%) | Rash | 1 (1.7%) | 1 (1.7%) due to rash |
Amenorrhea | 2 (3.4%) |
Herpes zoster | 1 (1.7%) |
Cystitis | 3 (5.2%) |
Pneumonia | 1 (1.7%) |
Hypotension | 1 (1.7%) |
Fatigue | 1 (1.7%) |
Mild hair loss | 4 (6.9%) |
Mealy et al. 2014 [26] | 28 | NR | NR | NR | 0 (0%) |
Torres et al. 2015 [20] | 11 | 4 (36%) | Sun sensitivity | NR | NR |
Recurrent infection | NR |
Chen et al. 2016 [13] | 62 | 3 (4.8%) | Mild hair loss | 2 (3.2%) | 0 (0%) |
Mildly elevated liver enzyme (After reused, no elevated liver enzyme) | 1 (1.6%) |
Jeong et al. 2016 [15] | 34 | NR | NR | NR | 0 (0%) |
Xu et al. 2016 [16] | 38 | 2 (5.3%) | Agranulocytosis | 1 (2.6%) | 2 (5.3%) due to agranulocytosis, amenorrhea |
Amenorrhea | 1 (2.6%) |
Chen et al. 2017 [14] | 105 | 5 (4.8%) | Mild hair loss | 3 (2.9%) | 0 (0%) |
Mildly elevated liver enzyme | 1 (1.0%) |
Phlegm on normal CT chest | 1 (1.0%) |
Montcuquet et al. 2017 [17] | 67 | 9 (13.4%) | Gastrointestinal side effects | 6 (9.0%) | 9 (13.4%) |
Infection | 3 (4.5%) |
Huang et al. 2018 [22] | 90 | 39 (43%) | Diarrhea | 2 (2.2%) | 8 (9%) |
Deranged liver enzyme | 18 (20%) |
Hyperbilirubinemia | 2 (2.2%) |
Respiratory infection | 11 (12.2%) |
Urinary tract infection | 5 (5.6%) |
Varicella-zoster virus infection | 5 (5.6%) |
Anemia | 6 (6.7%) |
Leukopenia | 4 (4.4%) |
Rectal cancer | 1 (1.1%) |
Renal insufficiency | 1 (1.1%) |
Hair loss | 2 (2.2%) |
Jiao et al. 2018 [12] | 109a | 21 (19%) | Hair loss | 5 (4.6%) | 1 (0.9%) |
Mildly elevated liver enzyme | 3 (2.8%) |
Diarrhea and abdominal pain | 2 (1.8%) |
Constipation | 1 (0.9%) |
Leukopenia | 3 (2.8%) |
Thrombocytopenia | 1 (0.9%) |
Shingles | 2 (1.8%) |
Herpes simplex infection | 2 (1.8%) |
Headache | 2 (1.8%) |
Chronic dermopathy of hands and nail | 1 (0.9%) |
Mealy et al. 2018 [21] | 245 | NR | NR | NR | NR |
Yang et al. 2018 [18] | 30 | 3 (10%) | Mildly elevated liver enzyme | 2 (6.7%) | 0 (0%) |
Nausea | 1 (3.3%) |
Zhou et al., 2019 [23] | 31 | NR | NR | NR | NR |
Poupart et al. 2020 [19] | 42 | 5 (11.9%)b | Serious infection events | 5 (11.9%) | 5 (11.9%) due to thrombocytopenia, arthromyalgia, Gastrointestinal side effects |
aTotal number of patients = 109 (86 of them received MMF > 6 months and were included in efficacy assessment) |
bThe article did not report adverse events other than serious infection events |
Abbreviation: NR, not reported |