- Email: [email protected]
Analysis of the treatment of neuromyelitis optica
Journal of the Neurological Sciences 351 (2015) 31–35
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Analysis of the treatment of neuromyelitis optica Jose Torres a,⁎, Amy Pruitt b, Laura Balcer a, Steven Galetta a, Clyde Markowitz b, Nabila Dahodwala b a b
Department of Neurology, New York University Langone Medical Center, USA Department of Neurology, University of Pennsylvania, USA
a r t i c l e
i n f o
Article history: Received 25 October 2014 Received in revised form 16 January 2015 Accepted 5 February 2015 Available online 13 February 2015 Keywords: NMO Neuromyelitis optica Devic's syndrome Treatment Rituximab Mycophenolate Azathioprine
a b s t r a c t Background: Treatment options for neuromyelitis optica (NMO) are currently based on small retrospective case series and open label studies, ranging from 10 to 103 patients. Objective: To compare the efficacy and tolerability of azathioprine, cyclophosphamide, mycophenolate, and rituximab in patients with neuromyelitis optica. Methods: This is a retrospective chart review and telephone follow-up study of 71 patients with NMO or NMO spectrum disorder, 54 of whom were treated with the study drugs. We compared adverse events, annualized relapse rates and expanded disability status scales before and after treatment. Results: The median ARR decreased from 1.17 to 0.25 on rituximab (P b 0.01), 0.92 to 0.56 on azathioprine (P = 0.475), 1.06 to 0.39 on mycophenolate (P b 0.05) and 1.30 to 0.92 on cyclophosphamide (P = 0.746). When compared directly to azathioprine, rituximab significantly reduced relapse rates (P = 0.021). The median EDSS decreased from 7 to 5 on rituximab (P b 0.01) and 7 to 6 on azathioprine (P b 0.01), and did not change significantly on mycophenolate (4 to 5; P = 0.463) or cyclophosphamide (6.5 to 6.5; P = 0.881). Twenty-five percent of patients noted adverse events on rituximab, 36% on azathioprine, 36% on mycophenolate, and 80% on cyclophosphamide. Conclusion: Rituximab significantly reduces relapse rates and improves disability while maintaining comparable tolerability to other immunosuppressive treatments for NMO. © 2015 Elsevier B.V. All rights reserved.
1. Introduction First described by Dr. Eugene Devic in 1894, neuromyelitis optica (NMO) is a rare demyelinating disorder of the central nervous system which primarily targets the optic nerves and the spinal cord. It is manifested by severe recurrent relapses and significant disability [1–3]. NMO was initially believed to be a variant of multiple sclerosis (MS) but after the discovery of the NMO IgG antibody in 2004, it has been reclassified as a separate disease entity [4]. While its pathogenesis has not been fully elucidated, recent evidence indicates that the disease may be due to a Bcell mediated process targeting aquaporin on astrocyte foot processes [5]. Immunosuppression is the mainstay of treatment [6–9,32]. However, there have been no randomized control trials to establish the efficacy of current treatment options. In fact, most of the data regarding the treatment of NMO are based on small retrospective case series and open label studies, ranging from 10 to 103 patients [10–28] (Table 1). Given this paucity of treatment data, we evaluated our experience in treating 54 patients with NMO at the Hospital of the University of ⁎ Corresponding author at: NYU Langone Medical Center, 530 First Avenue, HCC Building Suite 5A, New York, NY 10016, USA. Tel.: +1 646 501 0252; fax: +1 646 754 9778. E-mail address: [email protected] (J. Torres).
http://dx.doi.org/10.1016/j.jns.2015.02.012 0022-510X/© 2015 Elsevier B.V. All rights reserved.
Pennsylvania (HUP) and compared treatment efficacy between different agents. 2. Methods 2.1. Ethics We obtained approval from the Institutional Review Board at HUP with a waiver of written informed consent to conduct this study. Participants who were contacted by phone gave verbal consent after explanation of the study protocol, risks and benefits. 2.2. Case identification We conducted a search of our electronic medical record systems for all patients with billing codes for MS, transverse myelitis or optic neuritis treated with the study immunosuppressants or with billing codes for NMO or Devic's Disease. We also reviewed the records of all patients tested for the NMO IgG antibody over the past 5 years (the time during which these data were followed). We included patients with at least 6 months of follow-up who were NMO antibody positive with a history of transverse myelitis or optic neuritis, also known as NMO Spectrum Disorder (NMOSD), or those that met 2006 revised criteria for NMO. Those criteria stipulated that patients must have a history of optic
32
J. Torres et al. / Journal of the Neurological Sciences 351 (2015) 31–35
Table 1 Published NMO treatment studies.
Mandler et al. [22] Cree et al. [15] Jacob et al. [16] Jacob et al. [17] Bichuetti et al. [11] Pellkofer et al. [23] Kim et al. [11] Bedi et al. [10] Bomprezzi et al. [12] Costanzi et al. [14] Kim et al. [18] Lindsey et al. [21] Pittock et al. [25] Kitley et al. [20] Ayzenberg et al. [24] Cabre et al. [13] Yaguchi et al. [26] Kim et al. [19] Yang et al. [27] Mealy et al. [28] Elsone et al. [29]
Study design
# of patients
Intervention
Follow-up (median)
Prospective Retrospective Retrospective Retrospective Retrospective Prospective Retrospective Retrospective Retrospective Retrospective Prospective Retrospective Prospective Retrospective Prospective Prospective Retrospective Retrospective Prospective Retrospective Retrospective
7 8 25 24 36 10 20 23 18 99 30 9 14 14 3 51 4 30 5 90 103
Azathioprine Rituximab Rituximab Mycophenolate Multiple Rituximab Mitoxantrone Rituximab Rituximab Azathioprine Rituximab Rituximab Eculizumab Methotrexate Tocilizumab Mitoxantrone Cyclophosphamide Rituximab Rituximab Multiple Azathioprine
18 months 12 months 19 months 28 months 47 months 30 months 41 months 32 months 18 months 22 months 24 months 40 months 12 months 15 months 18 months 12 months 15 months 60 months 12 months 23 months 18 months
neuritis and acute myelitis and at least 2 of the following 3 supportive criteria: MRI of the brain not meeting criteria for MS; a contiguous spinal cord MRI lesion extending over ≥3 vertebral segments; or NMO IgG sero-positivity [30]. 2.3. Data collection For each patient meeting the above criteria, we obtained data regarding 1) demographics; 2) baseline clinical information; 3) radiographic data; 4) treatment details including dosing; 5) clinical course including documented relapses and neurologic examinations during hospitalizations and follow-up outpatient visits; and 6) adverse events. We established disease severity by calculating annualized relapse rates (ARRs) and patients' expanded disability status scales (EDSSs) [31]. Both of these metrics have been used in the evaluation of disease severity in NMO patients in prior published data [10–29]. We extrapolated EDSS scores based on the neurologic examinations documented in our records. We supplemented medical record data with a telephone call to all patients in order to verify the accuracy of our data and to obtain additional follow-up information. 2.4. Analysis To detect a difference of 1.5 in the ARR and 1 in the EDSS with a power of 0.80 and alpha ≤ 0.05, we needed a sample size of 24 patients per treatment group. We used Chi Square/Fisher Exact test for categorical variables and Mann–Whitney U Test for continuous variables to make bivariate comparisons between the main outcome variables (ARR and EDSS) and patient and clinical features. A two-sided Pvalue b 0.05 was considered significant. 3. Results
modifying therapies for MS. We were unable to obtain treatment information on 3 (4%) patients. There were an additional forty patients who were identified as NMO IgG positive or were coded as having a diagnosis of NMO through the electronic medical record, but had no clinical follow-up data in our system and could not be reached by phone. Of the 71 patients, 47 (66%) were NMO IgG sero-positive, 12 (17%) of which were NMOSD patients, and 23 (32%) were NMO IgG seronegative (Table 2). One patient was never tested for the NMO IgG antibody but met 2006 revised criteria for NMO. The mean age of symptom onset was 38 years overall but was non-significantly higher for patients with NMOSD, which had a mean onset age of 44 (P = 0.13). Sixty patients were female and 11 were male. Fifty-two percent of patients were African American, 46% were Caucasian and 2% were Asian. When comparing the patients by NMO subtype, gender, and race, the only significant finding was that 64% of men and 27% of females were NMO IgG sero-negative (P = 0.032). Interestingly, NMOSD patients had a lower median ARR when compared to all patients overall (0.80 vs 1.1) and had a lower median worst EDSS (5.25 vs 7.0), but these differences did not reach statistical significance. 3.2. Efficacy Thirty-two patients received rituximab, 22 azathioprine, 11 mycophenolate, and 5 cyclophosphamide. Therefore, only the azathioprine and rituximab treatment groups were sufficiently powered to make any statistically significant clinical comparisons. There were no statistically significant differences between those two treatment groups with respect to demographics, NMO IgG seropositivity, median disease duration prior to treatment, median time from most recent exacerbation to treatment, use of immunosuppressants prior to treatment, pretreatment EDSS, pre-treatment ARR, and median follow-up time after treatment (Table 3). However, the use of concomitant immunosuppressants such as corticosteroids, plasma exchange and disease modifying therapies for multiple sclerosis was significantly higher in the azathioprine group when compared to the rituximab group (55% vs 13%, P b 0.01). Despite this, the median ARR was reduced from 1.17 to 0.25 on rituximab (P b 0.01), 0.92 to 0.56 on azathioprine (P = 0.475), 1.06 to 0.39 on mycophenolate (P b 0.05) and 1.30 to 0.92 on cyclophosphamide (P = 0.746) (Fig. 1). Forty-seven percent of patients on rituximab remained symptom free on treatment in comparison to 32% on azathioprine, 27% on mycophenolate, and 20% on cyclophosphamide. The median post-treatment EDSS was reduced from 7 to 5 on rituximab (P b 0.01) and 7 to 6 on azathioprine (P b 0.01). The post-treatment EDSS increased on mycophenolate from 4 to 5 (P = 0.463) and remained the same at 6.5 on cyclophosphamide (P = 0.881) (Fig. 2). When compared directly with azathioprine, rituximab was more effective in reducing ARR (P = 0.021), but there was no statistically significant difference in its effects on EDSS. There was no significant effect of NMO IgG sero-positivity of rituximab dosing (usually either two doses of 1000 mg given one week apart or 375 mg/m2 given weekly for 4 weeks) on relapse rates and EDSS. Twenty-two of 32 patients received additional doses of rituximab only when levels of CD19 positive cells
Table 2 Patient demographics by NMO subtype.
3.1. Baseline demographic and clinical data We screened over 800 patients and 71 patients met criteria for NMO (59 patients) and NMOSD (12 patients) and had at least 6 months of clinical follow-up. Fifty-four (77%) of these patients were treated during the follow-up period with one or more of the study medications. Thirtytwo (45%) patients received rituximab, 22 (31%) azathioprine, 11 (15%) mycophenolate, and 5 (7%) cyclophosphamide. Eight (11%) patients never received treatment and 6 (8%) patients were treated with disease
Mean onset age Gender M F Ethnicity Caucasian African American Asian ⁎ p = 0.032
Total (N = 71)
NMO Ab+ (N = 47)
NMO Ab− (N = 23)
NMOSD (N = 12)
38 (12–71)
39 (12–71)
37 (18–62)
44 (13–70)
11 (14%) 60 (86%)
4 (9%) 43 (91%)
7 (30%)⁎ 16 (70%)
1 (8%) 11 (92%)
32 (45%) 37 (52%) 2 (3%)
22 (46%) 23 (48%) 2 (4%)
10 (43%) 13 (57%) 0
5 (42%) 6 (50%) 1 (8%)
J. Torres et al. / Journal of the Neurological Sciences 351 (2015) 31–35
33
Table 3 Demographics by treatment.
Mean onset age Gender M F Ethnicity Caucasian African American Asian NMO Ab+ Median time from most recent exacerbation to treatment (months) Median disease duration prior to treatment (months) Median follow-up time (months) Other immunosuppressants before treatment* Other immunosuppressants during treatment* Median pre-Tx ARR Median pre-Tx EDSS Reached by phone
Rituximab (N = 32)
Azathioprine (N = 22)
Mycophenolate (N = 11)
Cyclophosphamide (N = 5)
38 yrs (12–71 yrs)
39 yrs (13–68 yrs)
37 yrs (18–68 yrs)
35 yrs (21–53)
11 (14%) 60 (86%)
3 (9%) 29 (91%)
4 (18%) 18 (82%)
1 (9%) 10 (91%)
32 (45%) 37 (52%) 2 (3%) 23 (72%) 0 (0–3) 17.5 (5–69.5) 22 (14–39) 12 (38%) 4 (13%)⁎⁎
16 (50%) 16 (50%) 0 14 (64%) 0 (0–5.5) 16 (5–108) 21 (12–46) 10 (45%) 12 (55%) 0.92 (0–1.5) 7.0 (5.0–7.5) 16 (73%)
14 (64%) 7 (32%) 1 (4%) 4 (36%) 0 (0–5) 28 (22–71) 23 (13–60) 7 (64%) 5 (45%) 1.06 (0.84–2.31) 4.0 (3.0–6.5) 7 (64%)
4 (36%) 5 (46%) 2 (18%) 1 (20%) 0 (0–0) 6 (3–6) 19 (13–21) 1 (20%) 1 (20%) 1.30 (0–3.5) 6.5 (3.0–6.5) 4 (80%)
1.17 (0.77–3.66) 7.0 (5.5–7.25) 25 (78%)
⁎⁎ The azathioprine group had significantly more patients on concomitant immunosuppressive therapy when compared directly to the rituximab group (p b 0.01).
were greater than 2%, 3 patients received doses every 6 months regardless of symptoms and CD19 positive cell levels, and 7 patients received rituximab after clinical exacerbations. The samples were too small to establish any statistically significant differences.
3.3. Adverse events Twenty-five percent of patients noted adverse events on rituximab, 36% on azathioprine, 36% on mycophenolate, and 80% on cyclophosphamide. The most common complaints were leukopenia, recurrent infections, and nausea for azathioprine; pancytopenia, hair loss and nausea for cyclophosphamide; sun sensitivity and recurrent infections for mycophenolate; and leukopenia, recurrent infections and bacteremia/sepsis for rituximab. There were 5 deaths in total. All 5 patients died of sepsis thought secondary to wounds associated with immobility from paraplegia. At the time of death, one patient was on rituximab, one patient on azathioprine, and three were on no immunosuppressive therapy.
Fig. 1. Pre and post-treatment median annualized relapse rates (ARR). The median ARR decreased from 1.17 to 0.25 on rituximab (P b 0.01), 0.92 to 0.56 on azathioprine (P = 0.475), 1.06 to 0.39 on mycophenolate (P b 0.05) and 1.30 to 0.92 on cyclophosphamide (P = 0.746). When compared directly to azathioprine, rituximab significantly reduced ARR (P = 0.021).
4. Conclusion Our experience demonstrates the efficacy of rituximab in reducing ARR in patients with NMO and NMOSD with about half of the patients becoming relapse free on treatment. However, our absolute ARR reduction of 0.92 is lower than that of previously reported studies which demonstrated reductions between 1.3 and 2.6 [10,12,15,16,19,21,23, 27,28]. There are a few possible reasons for this difference. Many studies evaluated smaller cohorts than we did, which could increase the variability of their estimates [12,15,23,27]. In studies that had similar numbers of patients on rituximab, Kim et al. reported a median absolute ARR reduction of 2.1 both at 24 months and then at 60 months [19], Bedi et al. reported a median reduction of 1.87 [10], Mealy et al. reported a median reduction of 2.56 [28] and Jacob et al. reported a median reduction of 1.7 [16]. There was significant variability between the studies in terms of rituximab dosing protocols, patient demographics, and followup prior to and after treatment. All of these factors may have affected the reported median absolute ARR reductions. Moreover, while these studies have comparable numbers to ours, the number of patients analyzed in all studies is still small and there may be other unmeasured confounders and biases that could contribute to the differences across studies.
Fig. 2. Pre and post-treatment expanded disability status scales (EDSS). The median EDSS was reduced from 7 to 5 on rituximab (P b 0.01) and 7 to 6 on azathioprine (P b 0.01), remained the same at 6.5 on cyclophosphamide (P = 0.881), and increased from 4 to 5 (P = 0.463) on mycophenolate.
34
J. Torres et al. / Journal of the Neurological Sciences 351 (2015) 31–35
The treatment effect of rituximab appears to be important not only in stopping disease progression but also in allowing patients to heal as EDSS improved from a median of 7 to 5 with 11 patients experiencing an improvement in EDSS score of more than 3 points. This effect was statistically significant when compared directly with azathioprine and is in line with studies by Jacob et al. [16] and Bedi et al. [10]. However, Kim et al. reported a reduction in EDSS from 4.0 to 3.0 [19]. This sample was less disabled at the initiation of therapy which may partly explain the smaller effect. Additionally, other studies in Asian cohorts show similar reductions in EDSS. Yang et al. [27] looked at reduced-dose rituximab and Kim et al. [18] looked at mitoxantrone. They noted EDSS reductions of 4.5 to 4 and 5.6 to 4 respectively, indicating that perhaps Asian cohorts may have milder disease. Our results for the treatment efficacy of azathioprine are also different from prior published data. The most comparable studies to date with large cohorts of patients are by Costanzi et al. and Elsone et al. [14,29]. In the Costanzi study, the ARR was reduced from 2.20 to 0.89 and EDSS remained unchanged at 3.5, while in the Elsone paper, the median ARR decreased from 1.5 to 0 while EDSS improved from 6 to 5. This is different from our absolute ARR reduction of 0.36, which did not reach statistical significance. A possible explanation for the discrepant findings is likely related to the larger sample sizes of both studies and resulting increased power to detect a significant difference. Despite potential savings worth tens of thousands of dollars in hospitalizations, imaging, and rehabilitation costs associated with acute exacerbations, there are several major concerns regarding the use of rituximab including its safety, costs, and administration. Rituximab requires intravenous infusions and is therefore, more invasive. Four patients had to be admitted to our hospital with sepsis due to bacteremia on rituximab, whereas the number for azathioprine was only 1. Moreover, rituximab, is much more expensive with an average cost of about $27,000 a year (for four 1000 mg infusions) versus $2100 for azathioprine [14]. These concerns are particularly important when one considers that 32% of patients on azathioprine remained symptom free. Therefore, additional prospective studies will be needed in order to better assess the cost effectiveness of different therapies. This retrospective study had several limitations. First, our overall numbers were small and two of the treatment groups did not have enough patients to make any clinically significant comparisons. Second, this was not a randomized control trial so we cannot exclude selection bias in treatment choice. We looked at variables that may affect efficacy such as duration of disease before treatment, follow-up time after treatment, pre-treatment and concurrent immunosuppressant therapy and pre-treatment ARR and EDSS. The only significant difference was in concurrent immunosuppressive therapy and that was lowest in the rituximab group. These are, however, imperfect surrogate markers that may not fully account for possible selection bias. Third, there was large variability in pre-treatment disease duration ranging from 6 months for cyclophosphamide to 28 months for mycophenolate. This likely affected pre-treatment relapse rates and may have accounted for lower absolute ARR reductions within the treatment groups. Fourth, most of the patients in our study were initially seen in the hospital after acute exacerbations. Therefore, their initial EDSS was obtained during an exacerbation while follow-up EDSS was often obtained in the outpatient setting. This likely led to higher pre-treatment EDSS numbers and because the natural history of NMO is for partial recovery after an exacerbation, likely played a factor in the significant reduction in EDSS in the rituximab and azathioprine groups. This was not the case in the mycophenolate group where 6 of 11 patients had their pre-treatment EDSS calculated around the time of an exacerbation in an outpatient setting, presumably because they were less disabled. This may explain the lower pre-treatment EDSS in that group. Fifth, we did not analyze CD19 positive cell levels in the rituximab patients to see how effective the medication was at suppressing the patients' immune systems. Thus, while rituximab appeared to be an effective immunosuppressant, we do not have direct evidence of this immunosuppressive effect. Also,
as mentioned above, there were over forty patients who were NMO IgG positive or were coded as having a diagnosis of NMO but had no clinical data accessible in our system and could not be reached by phone. It is unclear if these patients met 2006 revised criteria for NMO or if they were lost to follow-up because of poor or good outcomes. Finally, this study also depended greatly on accurate documentation of patient events and exams. Although we tried to minimize the effect of poor documentation by calling patients directly, they often could not remember their histories in detail and may have given us biased answers when asked about the extent of their disability. Our cohort had a control group of sorts. There were 14 patients who never received study medications. Eight of these never received treatment and 6 were treated with disease modifying multiple sclerosis medications. The extent to which these patients are representative of all NMO patients remains unknown because most of these patients were either stable on MS medications that were started when NMO was thought to be a variant of MS or had one severe relapse without further relapses. This begs the question of why these untreated patients did not progress in the typically aggressive manner that we usually see with NMO and suggests further unexplored heterogeneity in the NMO population. Furthermore, if there is a subgroup of patients that do not have clinically progressive disease, it stands to reason that some of the treatment effect seen in our patients may have been because some of them would have been clinically stable with or without treatment. Given the potential costs and harm of long-term immunosuppressive treatment, it will be important to find out how to distinguish between patients that have progressive disease and those that do not. A randomized control trial is clearly needed in order to address the many questions that still exist regarding the clinical progression and treatment of NMO. However, based on the information from our study, it appears that rituximab is an effective and well tolerated treatment option for patients with NMO and NMOSD. Conflict of interest There are no conflicts of interest. References [1] Oh J, Levy M. Neuromyelitis optica: an antibody-mediated disorder of the central nervous system. Neurol Res Intl 2012;2012:1–13. [2] Mata S, Lolli F. Neuromyelitis optica: an update. J Neurol Sci Apr 15 2011;303(1–2): 13–21. [3] Wingerchuk DM, Hogancamp WF, O'Brien PC, Weinshenker BG. The clinical course of neuromyelitis optica (Devic’s syndrome). Neurology 1999;53:1107–14. [4] Lennon VA, Wingerchuk DM, Kryzer TJ, Pittock SJ, Lucchinetti CF, Fujihara K, et al. A serum autoantibody marker of neuromyelitis optica: distinction from multiple sclerosis. Lancet 2004;364:2106–12. [5] Bukhari W, Barnett MH, Prain K, Broadley SA. Molecular pathogenesis of neuromyelitis optica. Int J Mol Sci Oct 11 2012;13(10):12970–93. [6] Papeix C, Vidal JS, de Seze J, Pierrot-Deseilligny C, Tourbah A, Stankoff B, et al. Immunosuppressive therapy is more effective than interferon in neuromyelitis optica. Mult Scler 2007;13:256–9. [7] Collongues N, de Seze J. Current and future treatment approaches for neuromyelitis optica. Ther Adv Neurol Disord 2011;4:111–21. [8] Okamoto T, Ogawa M, Lin Y, Murata M, Miyake S, Yamamura T. Treatment of neuromyelitis optica: current debate. Ther Adv Neurol Disord Jul 2008;1(1):5–12. [9] Sato D, Callegaro D, Lana-Peixoto MA, Fujihara K. Treatment of neuromyelitis optica: an evidence based review. Neuropsiquiatria Jan 2012;70(1):59–66. [10] Bedi GS, Brown AD, Delgado SR, Usmani N, Lam BL, Sheremata WA. Impact of rituximab on relapse rate and disability in neuromyelitis optica. Mult Scler Oct 2011; 17(10):1225–30. [11] Bichuetti DB, Lobato de Oliveira EM, Oliveira DM, Amorin de Souza N, Gabbai AA. Neuromyelitis optica treatment: analysis of 36 patients. Arch Neurol 2010;67(9): 1131–6. [12] Bomprezzi R, Postevka E, Campagnolo D, Vollmer TL. A review of cases of neuromyelitis optica. Neurologist 2011;17(2):98–104. [13] Cabre P, Olindo S, Marignier R, Jeannin S, Merle H, Smadja D. Efficacy of mitoxantrone in neuromyelitis optica spectrum: clinical and neuroradiological study. J Neurol Neurosurg Psychiatry May 2013;84(5):511–6. [14] Costanzi C, Matiello M, Lucchinetti CF, Weinshenker BG, Pittock SJ, Mandrekar J, et al. Azathioprine: Tolerability, efficacy, and predictors of benefit in neuromyelitis optica. Neurology 2011;77(7):659–66.
J. Torres et al. / Journal of the Neurological Sciences 351 (2015) 31–35 [15] Cree BA, Lamb S, Morgan K, Chen A, Waubant E, Genain C. An open label study of the effects of rituximab in neuromyelitis optica. Neurology 2005;64(7):1270–2. [16] Jacob A, Weinshenker BG, Violich I, McLinskey N, Krupp L, Fox RJ, et al. Treatment of neuromyelitis optical with rituximab: Retrospective analysis of 25 patients. Arch Neurol 2008;65(11):1443–8. [17] Jacob A, Matiello M, Weinshenker BG, Wingerchuk DM, Lucchinetti C, Shuster E, et al. Treatment of neuromyelitis optica with mycophenolate mofetil: retrospective analysis of 24 patients. Arch Neurol 2009;66:1128–33. [18] Kim SH, Kim W, Park MS, Sohn EH, Li XF, Kim HJ. Efficacy and safety of mitoxantrone in patients with highly relapsing neuromyelitis optica. Arch Neurol 2011;68(4): 473–9. [19] Kim SH, Huh SY, Lee SJ, Joung A, Kim HJ. A 5 year follow up of rituximab treatment in patients with neuromyelitis optica spectrum disorder. JAMA Neurol 2013;70(9): 1110–7. [20] Kitley J, Elsone L, George J, Waters P, Woodhall M, Vincent A, et al. Methotrexate is an alternative to azathioprine in neuromyelitis optica spectrum disorders with aquaporin-4 antibodies. J Neurol Neurosurg Psychiatry Aug 2013;84(8):918–21. [21] Lindsey JW, Meulmester KM, Brod SA, Nelson F, Wolinsky JS. Variable results after rituximab in neuromyelitis optica. J Neurol Sci Jun 15 2012;317(1–2):103–5. [22] Mandler RN, Ahmed W, Dencoff JE. Devic’s neuromyelitis optica: a prospective study of seven patients treated with prednisone and azathioprine. Neurology 1998;51(4): 1219–20. [23] Pellkofer HL, Krumbholz M, Berthele A, Hemmer B, Gerdes LA, Havla J, et al. Longterm follow-up of patients with neuromyelitis optica after repeated therapy with rituximab. Neurology 2011;76:1310–5. [24] Ayzenberg I, Kleiter I, Schröder A, Hellwig K, Chan A, Yamamura T, et al. Interleukin 6 receptor blockade in patients with neuromyelitis optica nonresponsive to anti-CD20 therapy. JAMA Neurol Mar 1 2013;70(3):394–7.
35
[25] Pittock SJ, Lennon VA, McKeon A, Mandrekar J, Weinshenker BG, Lucchinetti CF, et al. Eculizumab in AQP4-IgG-positive relapsing neuromyelitis optica spectrum disorders: an open-label pilot study. Lancet Neurol Jun 2013;12(6):554–62. [26] Yaguchi H, Sakushima K, Takahashi I, Nishimura H, Yashima-Yamada M, Nakamura M, et al. Efficacy of intravenous cyclophosphamide therapy for neuromyelitis optica spectrum disorder. Intern Med 2013;52(9):969–72. [27] Yang CS, Yang L, Li T, Zhang DQ, Jin WN, Li MS, et al. Responsiveness to reduced dosage of rituximab in Chinese patients with neuromyelitis optica. Neurology 2013;81: 710–3. [28] Mealy MA, Wingerchuk DM, Palace J, Greenberg BM, Levy M. Comparison of relapse and treatment failure rates among patients with neuromyelitis optica: multicenter study of treatment efficacy. JAMA Neurol Mar 2014;71(3):324–30. [29] Elsone L, Kitley J, Luppe S, Lythgoe D, Mutch K, Jacob S, et al. Long-term efficacy, tolerability and retention rate of azathioprine in 103 aquaporin-4 antibody-positive neuromyelitis optica spectrum disorder patients: a multicentre retrospective observational study from the UK. Mult Scler Oct 2014;20(11):1533–40. [30] Wingerchuk DM, Lennon VA, Pittock SJ, Lucchinetti CF, Weinshenker BG. Revised diagnostic criteria for neuromyelitis optica. Neurology 2006;66(10):1485–9. [31] Kurtzke JF. Rating neurological impairment in multiple sclerosis: an expanded disability status scale. Neurology Nov 1983;33(11):1444–52. [32] Mealy MA, Wingerchuk DM, Greenberg BM, Levy M. Epidemiology of neuromyelitis optica in the United States: a multicenter analysis. Arch Neurol Sep 2012;69(9): 1176–80.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Analysis of the treatment of neuromyelitis optica Jose Torres a,⁎, Amy Pruitt b, Laura Balcer a, Steven Galetta a, Clyde Markowitz b, Nabila Dahodwala b a b
Department of Neurology, New York University Langone Medical Center, USA Department of Neurology, University of Pennsylvania, USA
a r t i c l e
i n f o
Article history: Received 25 October 2014 Received in revised form 16 January 2015 Accepted 5 February 2015 Available online 13 February 2015 Keywords: NMO Neuromyelitis optica Devic's syndrome Treatment Rituximab Mycophenolate Azathioprine
a b s t r a c t Background: Treatment options for neuromyelitis optica (NMO) are currently based on small retrospective case series and open label studies, ranging from 10 to 103 patients. Objective: To compare the efficacy and tolerability of azathioprine, cyclophosphamide, mycophenolate, and rituximab in patients with neuromyelitis optica. Methods: This is a retrospective chart review and telephone follow-up study of 71 patients with NMO or NMO spectrum disorder, 54 of whom were treated with the study drugs. We compared adverse events, annualized relapse rates and expanded disability status scales before and after treatment. Results: The median ARR decreased from 1.17 to 0.25 on rituximab (P b 0.01), 0.92 to 0.56 on azathioprine (P = 0.475), 1.06 to 0.39 on mycophenolate (P b 0.05) and 1.30 to 0.92 on cyclophosphamide (P = 0.746). When compared directly to azathioprine, rituximab significantly reduced relapse rates (P = 0.021). The median EDSS decreased from 7 to 5 on rituximab (P b 0.01) and 7 to 6 on azathioprine (P b 0.01), and did not change significantly on mycophenolate (4 to 5; P = 0.463) or cyclophosphamide (6.5 to 6.5; P = 0.881). Twenty-five percent of patients noted adverse events on rituximab, 36% on azathioprine, 36% on mycophenolate, and 80% on cyclophosphamide. Conclusion: Rituximab significantly reduces relapse rates and improves disability while maintaining comparable tolerability to other immunosuppressive treatments for NMO. © 2015 Elsevier B.V. All rights reserved.
1. Introduction First described by Dr. Eugene Devic in 1894, neuromyelitis optica (NMO) is a rare demyelinating disorder of the central nervous system which primarily targets the optic nerves and the spinal cord. It is manifested by severe recurrent relapses and significant disability [1–3]. NMO was initially believed to be a variant of multiple sclerosis (MS) but after the discovery of the NMO IgG antibody in 2004, it has been reclassified as a separate disease entity [4]. While its pathogenesis has not been fully elucidated, recent evidence indicates that the disease may be due to a Bcell mediated process targeting aquaporin on astrocyte foot processes [5]. Immunosuppression is the mainstay of treatment [6–9,32]. However, there have been no randomized control trials to establish the efficacy of current treatment options. In fact, most of the data regarding the treatment of NMO are based on small retrospective case series and open label studies, ranging from 10 to 103 patients [10–28] (Table 1). Given this paucity of treatment data, we evaluated our experience in treating 54 patients with NMO at the Hospital of the University of ⁎ Corresponding author at: NYU Langone Medical Center, 530 First Avenue, HCC Building Suite 5A, New York, NY 10016, USA. Tel.: +1 646 501 0252; fax: +1 646 754 9778. E-mail address: [email protected] (J. Torres).
http://dx.doi.org/10.1016/j.jns.2015.02.012 0022-510X/© 2015 Elsevier B.V. All rights reserved.
Pennsylvania (HUP) and compared treatment efficacy between different agents. 2. Methods 2.1. Ethics We obtained approval from the Institutional Review Board at HUP with a waiver of written informed consent to conduct this study. Participants who were contacted by phone gave verbal consent after explanation of the study protocol, risks and benefits. 2.2. Case identification We conducted a search of our electronic medical record systems for all patients with billing codes for MS, transverse myelitis or optic neuritis treated with the study immunosuppressants or with billing codes for NMO or Devic's Disease. We also reviewed the records of all patients tested for the NMO IgG antibody over the past 5 years (the time during which these data were followed). We included patients with at least 6 months of follow-up who were NMO antibody positive with a history of transverse myelitis or optic neuritis, also known as NMO Spectrum Disorder (NMOSD), or those that met 2006 revised criteria for NMO. Those criteria stipulated that patients must have a history of optic
32
J. Torres et al. / Journal of the Neurological Sciences 351 (2015) 31–35
Table 1 Published NMO treatment studies.
Mandler et al. [22] Cree et al. [15] Jacob et al. [16] Jacob et al. [17] Bichuetti et al. [11] Pellkofer et al. [23] Kim et al. [11] Bedi et al. [10] Bomprezzi et al. [12] Costanzi et al. [14] Kim et al. [18] Lindsey et al. [21] Pittock et al. [25] Kitley et al. [20] Ayzenberg et al. [24] Cabre et al. [13] Yaguchi et al. [26] Kim et al. [19] Yang et al. [27] Mealy et al. [28] Elsone et al. [29]
Study design
# of patients
Intervention
Follow-up (median)
Prospective Retrospective Retrospective Retrospective Retrospective Prospective Retrospective Retrospective Retrospective Retrospective Prospective Retrospective Prospective Retrospective Prospective Prospective Retrospective Retrospective Prospective Retrospective Retrospective
7 8 25 24 36 10 20 23 18 99 30 9 14 14 3 51 4 30 5 90 103
Azathioprine Rituximab Rituximab Mycophenolate Multiple Rituximab Mitoxantrone Rituximab Rituximab Azathioprine Rituximab Rituximab Eculizumab Methotrexate Tocilizumab Mitoxantrone Cyclophosphamide Rituximab Rituximab Multiple Azathioprine
18 months 12 months 19 months 28 months 47 months 30 months 41 months 32 months 18 months 22 months 24 months 40 months 12 months 15 months 18 months 12 months 15 months 60 months 12 months 23 months 18 months
neuritis and acute myelitis and at least 2 of the following 3 supportive criteria: MRI of the brain not meeting criteria for MS; a contiguous spinal cord MRI lesion extending over ≥3 vertebral segments; or NMO IgG sero-positivity [30]. 2.3. Data collection For each patient meeting the above criteria, we obtained data regarding 1) demographics; 2) baseline clinical information; 3) radiographic data; 4) treatment details including dosing; 5) clinical course including documented relapses and neurologic examinations during hospitalizations and follow-up outpatient visits; and 6) adverse events. We established disease severity by calculating annualized relapse rates (ARRs) and patients' expanded disability status scales (EDSSs) [31]. Both of these metrics have been used in the evaluation of disease severity in NMO patients in prior published data [10–29]. We extrapolated EDSS scores based on the neurologic examinations documented in our records. We supplemented medical record data with a telephone call to all patients in order to verify the accuracy of our data and to obtain additional follow-up information. 2.4. Analysis To detect a difference of 1.5 in the ARR and 1 in the EDSS with a power of 0.80 and alpha ≤ 0.05, we needed a sample size of 24 patients per treatment group. We used Chi Square/Fisher Exact test for categorical variables and Mann–Whitney U Test for continuous variables to make bivariate comparisons between the main outcome variables (ARR and EDSS) and patient and clinical features. A two-sided Pvalue b 0.05 was considered significant. 3. Results
modifying therapies for MS. We were unable to obtain treatment information on 3 (4%) patients. There were an additional forty patients who were identified as NMO IgG positive or were coded as having a diagnosis of NMO through the electronic medical record, but had no clinical follow-up data in our system and could not be reached by phone. Of the 71 patients, 47 (66%) were NMO IgG sero-positive, 12 (17%) of which were NMOSD patients, and 23 (32%) were NMO IgG seronegative (Table 2). One patient was never tested for the NMO IgG antibody but met 2006 revised criteria for NMO. The mean age of symptom onset was 38 years overall but was non-significantly higher for patients with NMOSD, which had a mean onset age of 44 (P = 0.13). Sixty patients were female and 11 were male. Fifty-two percent of patients were African American, 46% were Caucasian and 2% were Asian. When comparing the patients by NMO subtype, gender, and race, the only significant finding was that 64% of men and 27% of females were NMO IgG sero-negative (P = 0.032). Interestingly, NMOSD patients had a lower median ARR when compared to all patients overall (0.80 vs 1.1) and had a lower median worst EDSS (5.25 vs 7.0), but these differences did not reach statistical significance. 3.2. Efficacy Thirty-two patients received rituximab, 22 azathioprine, 11 mycophenolate, and 5 cyclophosphamide. Therefore, only the azathioprine and rituximab treatment groups were sufficiently powered to make any statistically significant clinical comparisons. There were no statistically significant differences between those two treatment groups with respect to demographics, NMO IgG seropositivity, median disease duration prior to treatment, median time from most recent exacerbation to treatment, use of immunosuppressants prior to treatment, pretreatment EDSS, pre-treatment ARR, and median follow-up time after treatment (Table 3). However, the use of concomitant immunosuppressants such as corticosteroids, plasma exchange and disease modifying therapies for multiple sclerosis was significantly higher in the azathioprine group when compared to the rituximab group (55% vs 13%, P b 0.01). Despite this, the median ARR was reduced from 1.17 to 0.25 on rituximab (P b 0.01), 0.92 to 0.56 on azathioprine (P = 0.475), 1.06 to 0.39 on mycophenolate (P b 0.05) and 1.30 to 0.92 on cyclophosphamide (P = 0.746) (Fig. 1). Forty-seven percent of patients on rituximab remained symptom free on treatment in comparison to 32% on azathioprine, 27% on mycophenolate, and 20% on cyclophosphamide. The median post-treatment EDSS was reduced from 7 to 5 on rituximab (P b 0.01) and 7 to 6 on azathioprine (P b 0.01). The post-treatment EDSS increased on mycophenolate from 4 to 5 (P = 0.463) and remained the same at 6.5 on cyclophosphamide (P = 0.881) (Fig. 2). When compared directly with azathioprine, rituximab was more effective in reducing ARR (P = 0.021), but there was no statistically significant difference in its effects on EDSS. There was no significant effect of NMO IgG sero-positivity of rituximab dosing (usually either two doses of 1000 mg given one week apart or 375 mg/m2 given weekly for 4 weeks) on relapse rates and EDSS. Twenty-two of 32 patients received additional doses of rituximab only when levels of CD19 positive cells
Table 2 Patient demographics by NMO subtype.
3.1. Baseline demographic and clinical data We screened over 800 patients and 71 patients met criteria for NMO (59 patients) and NMOSD (12 patients) and had at least 6 months of clinical follow-up. Fifty-four (77%) of these patients were treated during the follow-up period with one or more of the study medications. Thirtytwo (45%) patients received rituximab, 22 (31%) azathioprine, 11 (15%) mycophenolate, and 5 (7%) cyclophosphamide. Eight (11%) patients never received treatment and 6 (8%) patients were treated with disease
Mean onset age Gender M F Ethnicity Caucasian African American Asian ⁎ p = 0.032
Total (N = 71)
NMO Ab+ (N = 47)
NMO Ab− (N = 23)
NMOSD (N = 12)
38 (12–71)
39 (12–71)
37 (18–62)
44 (13–70)
11 (14%) 60 (86%)
4 (9%) 43 (91%)
7 (30%)⁎ 16 (70%)
1 (8%) 11 (92%)
32 (45%) 37 (52%) 2 (3%)
22 (46%) 23 (48%) 2 (4%)
10 (43%) 13 (57%) 0
5 (42%) 6 (50%) 1 (8%)
J. Torres et al. / Journal of the Neurological Sciences 351 (2015) 31–35
33
Table 3 Demographics by treatment.
Mean onset age Gender M F Ethnicity Caucasian African American Asian NMO Ab+ Median time from most recent exacerbation to treatment (months) Median disease duration prior to treatment (months) Median follow-up time (months) Other immunosuppressants before treatment* Other immunosuppressants during treatment* Median pre-Tx ARR Median pre-Tx EDSS Reached by phone
Rituximab (N = 32)
Azathioprine (N = 22)
Mycophenolate (N = 11)
Cyclophosphamide (N = 5)
38 yrs (12–71 yrs)
39 yrs (13–68 yrs)
37 yrs (18–68 yrs)
35 yrs (21–53)
11 (14%) 60 (86%)
3 (9%) 29 (91%)
4 (18%) 18 (82%)
1 (9%) 10 (91%)
32 (45%) 37 (52%) 2 (3%) 23 (72%) 0 (0–3) 17.5 (5–69.5) 22 (14–39) 12 (38%) 4 (13%)⁎⁎
16 (50%) 16 (50%) 0 14 (64%) 0 (0–5.5) 16 (5–108) 21 (12–46) 10 (45%) 12 (55%) 0.92 (0–1.5) 7.0 (5.0–7.5) 16 (73%)
14 (64%) 7 (32%) 1 (4%) 4 (36%) 0 (0–5) 28 (22–71) 23 (13–60) 7 (64%) 5 (45%) 1.06 (0.84–2.31) 4.0 (3.0–6.5) 7 (64%)
4 (36%) 5 (46%) 2 (18%) 1 (20%) 0 (0–0) 6 (3–6) 19 (13–21) 1 (20%) 1 (20%) 1.30 (0–3.5) 6.5 (3.0–6.5) 4 (80%)
1.17 (0.77–3.66) 7.0 (5.5–7.25) 25 (78%)
⁎⁎ The azathioprine group had significantly more patients on concomitant immunosuppressive therapy when compared directly to the rituximab group (p b 0.01).
were greater than 2%, 3 patients received doses every 6 months regardless of symptoms and CD19 positive cell levels, and 7 patients received rituximab after clinical exacerbations. The samples were too small to establish any statistically significant differences.
3.3. Adverse events Twenty-five percent of patients noted adverse events on rituximab, 36% on azathioprine, 36% on mycophenolate, and 80% on cyclophosphamide. The most common complaints were leukopenia, recurrent infections, and nausea for azathioprine; pancytopenia, hair loss and nausea for cyclophosphamide; sun sensitivity and recurrent infections for mycophenolate; and leukopenia, recurrent infections and bacteremia/sepsis for rituximab. There were 5 deaths in total. All 5 patients died of sepsis thought secondary to wounds associated with immobility from paraplegia. At the time of death, one patient was on rituximab, one patient on azathioprine, and three were on no immunosuppressive therapy.
Fig. 1. Pre and post-treatment median annualized relapse rates (ARR). The median ARR decreased from 1.17 to 0.25 on rituximab (P b 0.01), 0.92 to 0.56 on azathioprine (P = 0.475), 1.06 to 0.39 on mycophenolate (P b 0.05) and 1.30 to 0.92 on cyclophosphamide (P = 0.746). When compared directly to azathioprine, rituximab significantly reduced ARR (P = 0.021).
4. Conclusion Our experience demonstrates the efficacy of rituximab in reducing ARR in patients with NMO and NMOSD with about half of the patients becoming relapse free on treatment. However, our absolute ARR reduction of 0.92 is lower than that of previously reported studies which demonstrated reductions between 1.3 and 2.6 [10,12,15,16,19,21,23, 27,28]. There are a few possible reasons for this difference. Many studies evaluated smaller cohorts than we did, which could increase the variability of their estimates [12,15,23,27]. In studies that had similar numbers of patients on rituximab, Kim et al. reported a median absolute ARR reduction of 2.1 both at 24 months and then at 60 months [19], Bedi et al. reported a median reduction of 1.87 [10], Mealy et al. reported a median reduction of 2.56 [28] and Jacob et al. reported a median reduction of 1.7 [16]. There was significant variability between the studies in terms of rituximab dosing protocols, patient demographics, and followup prior to and after treatment. All of these factors may have affected the reported median absolute ARR reductions. Moreover, while these studies have comparable numbers to ours, the number of patients analyzed in all studies is still small and there may be other unmeasured confounders and biases that could contribute to the differences across studies.
Fig. 2. Pre and post-treatment expanded disability status scales (EDSS). The median EDSS was reduced from 7 to 5 on rituximab (P b 0.01) and 7 to 6 on azathioprine (P b 0.01), remained the same at 6.5 on cyclophosphamide (P = 0.881), and increased from 4 to 5 (P = 0.463) on mycophenolate.
34
J. Torres et al. / Journal of the Neurological Sciences 351 (2015) 31–35
The treatment effect of rituximab appears to be important not only in stopping disease progression but also in allowing patients to heal as EDSS improved from a median of 7 to 5 with 11 patients experiencing an improvement in EDSS score of more than 3 points. This effect was statistically significant when compared directly with azathioprine and is in line with studies by Jacob et al. [16] and Bedi et al. [10]. However, Kim et al. reported a reduction in EDSS from 4.0 to 3.0 [19]. This sample was less disabled at the initiation of therapy which may partly explain the smaller effect. Additionally, other studies in Asian cohorts show similar reductions in EDSS. Yang et al. [27] looked at reduced-dose rituximab and Kim et al. [18] looked at mitoxantrone. They noted EDSS reductions of 4.5 to 4 and 5.6 to 4 respectively, indicating that perhaps Asian cohorts may have milder disease. Our results for the treatment efficacy of azathioprine are also different from prior published data. The most comparable studies to date with large cohorts of patients are by Costanzi et al. and Elsone et al. [14,29]. In the Costanzi study, the ARR was reduced from 2.20 to 0.89 and EDSS remained unchanged at 3.5, while in the Elsone paper, the median ARR decreased from 1.5 to 0 while EDSS improved from 6 to 5. This is different from our absolute ARR reduction of 0.36, which did not reach statistical significance. A possible explanation for the discrepant findings is likely related to the larger sample sizes of both studies and resulting increased power to detect a significant difference. Despite potential savings worth tens of thousands of dollars in hospitalizations, imaging, and rehabilitation costs associated with acute exacerbations, there are several major concerns regarding the use of rituximab including its safety, costs, and administration. Rituximab requires intravenous infusions and is therefore, more invasive. Four patients had to be admitted to our hospital with sepsis due to bacteremia on rituximab, whereas the number for azathioprine was only 1. Moreover, rituximab, is much more expensive with an average cost of about $27,000 a year (for four 1000 mg infusions) versus $2100 for azathioprine [14]. These concerns are particularly important when one considers that 32% of patients on azathioprine remained symptom free. Therefore, additional prospective studies will be needed in order to better assess the cost effectiveness of different therapies. This retrospective study had several limitations. First, our overall numbers were small and two of the treatment groups did not have enough patients to make any clinically significant comparisons. Second, this was not a randomized control trial so we cannot exclude selection bias in treatment choice. We looked at variables that may affect efficacy such as duration of disease before treatment, follow-up time after treatment, pre-treatment and concurrent immunosuppressant therapy and pre-treatment ARR and EDSS. The only significant difference was in concurrent immunosuppressive therapy and that was lowest in the rituximab group. These are, however, imperfect surrogate markers that may not fully account for possible selection bias. Third, there was large variability in pre-treatment disease duration ranging from 6 months for cyclophosphamide to 28 months for mycophenolate. This likely affected pre-treatment relapse rates and may have accounted for lower absolute ARR reductions within the treatment groups. Fourth, most of the patients in our study were initially seen in the hospital after acute exacerbations. Therefore, their initial EDSS was obtained during an exacerbation while follow-up EDSS was often obtained in the outpatient setting. This likely led to higher pre-treatment EDSS numbers and because the natural history of NMO is for partial recovery after an exacerbation, likely played a factor in the significant reduction in EDSS in the rituximab and azathioprine groups. This was not the case in the mycophenolate group where 6 of 11 patients had their pre-treatment EDSS calculated around the time of an exacerbation in an outpatient setting, presumably because they were less disabled. This may explain the lower pre-treatment EDSS in that group. Fifth, we did not analyze CD19 positive cell levels in the rituximab patients to see how effective the medication was at suppressing the patients' immune systems. Thus, while rituximab appeared to be an effective immunosuppressant, we do not have direct evidence of this immunosuppressive effect. Also,
as mentioned above, there were over forty patients who were NMO IgG positive or were coded as having a diagnosis of NMO but had no clinical data accessible in our system and could not be reached by phone. It is unclear if these patients met 2006 revised criteria for NMO or if they were lost to follow-up because of poor or good outcomes. Finally, this study also depended greatly on accurate documentation of patient events and exams. Although we tried to minimize the effect of poor documentation by calling patients directly, they often could not remember their histories in detail and may have given us biased answers when asked about the extent of their disability. Our cohort had a control group of sorts. There were 14 patients who never received study medications. Eight of these never received treatment and 6 were treated with disease modifying multiple sclerosis medications. The extent to which these patients are representative of all NMO patients remains unknown because most of these patients were either stable on MS medications that were started when NMO was thought to be a variant of MS or had one severe relapse without further relapses. This begs the question of why these untreated patients did not progress in the typically aggressive manner that we usually see with NMO and suggests further unexplored heterogeneity in the NMO population. Furthermore, if there is a subgroup of patients that do not have clinically progressive disease, it stands to reason that some of the treatment effect seen in our patients may have been because some of them would have been clinically stable with or without treatment. Given the potential costs and harm of long-term immunosuppressive treatment, it will be important to find out how to distinguish between patients that have progressive disease and those that do not. A randomized control trial is clearly needed in order to address the many questions that still exist regarding the clinical progression and treatment of NMO. However, based on the information from our study, it appears that rituximab is an effective and well tolerated treatment option for patients with NMO and NMOSD. Conflict of interest There are no conflicts of interest. References [1] Oh J, Levy M. Neuromyelitis optica: an antibody-mediated disorder of the central nervous system. Neurol Res Intl 2012;2012:1–13. [2] Mata S, Lolli F. Neuromyelitis optica: an update. J Neurol Sci Apr 15 2011;303(1–2): 13–21. [3] Wingerchuk DM, Hogancamp WF, O'Brien PC, Weinshenker BG. The clinical course of neuromyelitis optica (Devic’s syndrome). Neurology 1999;53:1107–14. [4] Lennon VA, Wingerchuk DM, Kryzer TJ, Pittock SJ, Lucchinetti CF, Fujihara K, et al. A serum autoantibody marker of neuromyelitis optica: distinction from multiple sclerosis. Lancet 2004;364:2106–12. [5] Bukhari W, Barnett MH, Prain K, Broadley SA. Molecular pathogenesis of neuromyelitis optica. Int J Mol Sci Oct 11 2012;13(10):12970–93. [6] Papeix C, Vidal JS, de Seze J, Pierrot-Deseilligny C, Tourbah A, Stankoff B, et al. Immunosuppressive therapy is more effective than interferon in neuromyelitis optica. Mult Scler 2007;13:256–9. [7] Collongues N, de Seze J. Current and future treatment approaches for neuromyelitis optica. Ther Adv Neurol Disord 2011;4:111–21. [8] Okamoto T, Ogawa M, Lin Y, Murata M, Miyake S, Yamamura T. Treatment of neuromyelitis optica: current debate. Ther Adv Neurol Disord Jul 2008;1(1):5–12. [9] Sato D, Callegaro D, Lana-Peixoto MA, Fujihara K. Treatment of neuromyelitis optica: an evidence based review. Neuropsiquiatria Jan 2012;70(1):59–66. [10] Bedi GS, Brown AD, Delgado SR, Usmani N, Lam BL, Sheremata WA. Impact of rituximab on relapse rate and disability in neuromyelitis optica. Mult Scler Oct 2011; 17(10):1225–30. [11] Bichuetti DB, Lobato de Oliveira EM, Oliveira DM, Amorin de Souza N, Gabbai AA. Neuromyelitis optica treatment: analysis of 36 patients. Arch Neurol 2010;67(9): 1131–6. [12] Bomprezzi R, Postevka E, Campagnolo D, Vollmer TL. A review of cases of neuromyelitis optica. Neurologist 2011;17(2):98–104. [13] Cabre P, Olindo S, Marignier R, Jeannin S, Merle H, Smadja D. Efficacy of mitoxantrone in neuromyelitis optica spectrum: clinical and neuroradiological study. J Neurol Neurosurg Psychiatry May 2013;84(5):511–6. [14] Costanzi C, Matiello M, Lucchinetti CF, Weinshenker BG, Pittock SJ, Mandrekar J, et al. Azathioprine: Tolerability, efficacy, and predictors of benefit in neuromyelitis optica. Neurology 2011;77(7):659–66.
J. Torres et al. / Journal of the Neurological Sciences 351 (2015) 31–35 [15] Cree BA, Lamb S, Morgan K, Chen A, Waubant E, Genain C. An open label study of the effects of rituximab in neuromyelitis optica. Neurology 2005;64(7):1270–2. [16] Jacob A, Weinshenker BG, Violich I, McLinskey N, Krupp L, Fox RJ, et al. Treatment of neuromyelitis optical with rituximab: Retrospective analysis of 25 patients. Arch Neurol 2008;65(11):1443–8. [17] Jacob A, Matiello M, Weinshenker BG, Wingerchuk DM, Lucchinetti C, Shuster E, et al. Treatment of neuromyelitis optica with mycophenolate mofetil: retrospective analysis of 24 patients. Arch Neurol 2009;66:1128–33. [18] Kim SH, Kim W, Park MS, Sohn EH, Li XF, Kim HJ. Efficacy and safety of mitoxantrone in patients with highly relapsing neuromyelitis optica. Arch Neurol 2011;68(4): 473–9. [19] Kim SH, Huh SY, Lee SJ, Joung A, Kim HJ. A 5 year follow up of rituximab treatment in patients with neuromyelitis optica spectrum disorder. JAMA Neurol 2013;70(9): 1110–7. [20] Kitley J, Elsone L, George J, Waters P, Woodhall M, Vincent A, et al. Methotrexate is an alternative to azathioprine in neuromyelitis optica spectrum disorders with aquaporin-4 antibodies. J Neurol Neurosurg Psychiatry Aug 2013;84(8):918–21. [21] Lindsey JW, Meulmester KM, Brod SA, Nelson F, Wolinsky JS. Variable results after rituximab in neuromyelitis optica. J Neurol Sci Jun 15 2012;317(1–2):103–5. [22] Mandler RN, Ahmed W, Dencoff JE. Devic’s neuromyelitis optica: a prospective study of seven patients treated with prednisone and azathioprine. Neurology 1998;51(4): 1219–20. [23] Pellkofer HL, Krumbholz M, Berthele A, Hemmer B, Gerdes LA, Havla J, et al. Longterm follow-up of patients with neuromyelitis optica after repeated therapy with rituximab. Neurology 2011;76:1310–5. [24] Ayzenberg I, Kleiter I, Schröder A, Hellwig K, Chan A, Yamamura T, et al. Interleukin 6 receptor blockade in patients with neuromyelitis optica nonresponsive to anti-CD20 therapy. JAMA Neurol Mar 1 2013;70(3):394–7.
35
[25] Pittock SJ, Lennon VA, McKeon A, Mandrekar J, Weinshenker BG, Lucchinetti CF, et al. Eculizumab in AQP4-IgG-positive relapsing neuromyelitis optica spectrum disorders: an open-label pilot study. Lancet Neurol Jun 2013;12(6):554–62. [26] Yaguchi H, Sakushima K, Takahashi I, Nishimura H, Yashima-Yamada M, Nakamura M, et al. Efficacy of intravenous cyclophosphamide therapy for neuromyelitis optica spectrum disorder. Intern Med 2013;52(9):969–72. [27] Yang CS, Yang L, Li T, Zhang DQ, Jin WN, Li MS, et al. Responsiveness to reduced dosage of rituximab in Chinese patients with neuromyelitis optica. Neurology 2013;81: 710–3. [28] Mealy MA, Wingerchuk DM, Palace J, Greenberg BM, Levy M. Comparison of relapse and treatment failure rates among patients with neuromyelitis optica: multicenter study of treatment efficacy. JAMA Neurol Mar 2014;71(3):324–30. [29] Elsone L, Kitley J, Luppe S, Lythgoe D, Mutch K, Jacob S, et al. Long-term efficacy, tolerability and retention rate of azathioprine in 103 aquaporin-4 antibody-positive neuromyelitis optica spectrum disorder patients: a multicentre retrospective observational study from the UK. Mult Scler Oct 2014;20(11):1533–40. [30] Wingerchuk DM, Lennon VA, Pittock SJ, Lucchinetti CF, Weinshenker BG. Revised diagnostic criteria for neuromyelitis optica. Neurology 2006;66(10):1485–9. [31] Kurtzke JF. Rating neurological impairment in multiple sclerosis: an expanded disability status scale. Neurology Nov 1983;33(11):1444–52. [32] Mealy MA, Wingerchuk DM, Greenberg BM, Levy M. Epidemiology of neuromyelitis optica in the United States: a multicenter analysis. Arch Neurol Sep 2012;69(9): 1176–80.