If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Effectiveness of rituximab versus oral immunosuppressive therapies in neuromyelitis optica spectrum disorder in a racially diverse cohort of subjects: A single-center retrospective study
Neuromyelitis optica spectrum disorder (NMOSD) is an autoimmune, inflammatory disorder characterized by severe relapses and high level of disability. In clinical trials of NMOSD, Black patients are under-represented, < 12%, compared to a relatively high prevalence of NMSOD in this population, 10/100,000. Despite the higher prevalence of NMOSD in Black and Asian patients, there is limited knowledge of the effectiveness of disease modifying treatments across racially diverse groups.
Objective
To assess the effectiveness of rituximab and oral immunosuppressive agents in a cohort of NMOSD patients, the majority of whom are Black, in a real-world, clinical setting.
Methods
A single-center retrospective study was conducted at the University of Chicago Medical Center. Inclusion criteria: (1) diagnosis according to the 2015 International Panel for NMO Diagnosis (IPND) Criteria, (2) positive anti-aquaporin-4 antibodies on ELISA or cell-based tests, (3) initiation of immunosuppressant therapy within 5 years of disease onset, (4) first-line treatment with rituximab, mycophenolate (MMF), or azathioprine (AZA). Patients with negative anti-AQP4 titers were excluded. Kaplan-Meier survival analysis was used to estimate proportion of relapse-free patients following initiation of first line immunosuppressive therapy. A Cox proportional hazards regression model assessed the association of risk of relapsing with first-line immunosuppressive treatments with and without adjustments of pre-specified factors (age at disease onset, duration of disease, sex, race, CNS location of relapse).
Results
7 of 29 patients (24%) receiving rituximab experienced a relapse within the first 3 years of treatment vs. 13 of 23 patients (57%) receiving either AZA or MMF. Within the first 6 months of treatment, 2 (6.9%) patients treated with rituximab experienced a relapse vs. 7 (30.4%) patients treated with either MMF or AZA. In the 29 patients treated with rituximab, the 1-year and 3-year proportion of relapse-free patients was 88.8% and 70.9%. For the 23 patients treated with either AZA or MMF, the 1-year and 3-year proportion of relapse-free patients was 69.5% and 38.7%. In the univariate analysis, the risk of relapse was significantly higher in patients treated with AZA or MMF compared to those treated with rituximab (hazard ratio [HR] of 2.48 [0.99 - 6.21]; p = 0.046).
Conclusion
In this real-world study involving a majority of Black NMOSD patients, rituximab was relatively more effective in preventing relapses within 3 years of therapy initiation than AZA and MMF. Rituximab remains an effective option for treating NMOSD, especially when there are delays in treatment due to access and economic issues associated with newer treatments.
Neuromyelitis optica spectrum disorder (NMOSD) is a severe, autoimmune, inflammatory disorder of the central nervous system (CNS) characterized by longitudinally extensive lesions in the spinal cord and the optic nerves, although the brainstem and supratentorial regions can also be involved. The pathological hallmark of NMOSD is astrocytic damage mediated by anti-aquaporin 4 antibody (anti-AQP4) (
), which can penetrate the blood brain barrier and bind to the foot processes of astrocytes, triggering both a destructive, immune cascade (complement mediated) and local astrocytic toxicity through antibody binding and internalization, i.e., sublytic astrocytopathy (
). In addition to the pathological role of the AQP4 antibody produced by B cells, other mediators of inflammation such as IL-6, neutrophils, eosinophils, macrophages, and microglia play a prominent role in injury, and are potential targets for disease modification (
). Relapses in NMOSD are associated with severe inflammation, resulting in greater tissue injury and worse recovery than seen in the demyelinating diseases, such as multiple sclerosis. Given the high disability ensuing from NMOSD relapses, accurate diagnosis and initiation of effective treatment soon after the first relapse improves prognosis.
Rituximab, mycophenolate mofetil (MMF), and azathioprine (AZA) have traditionally been used as first-line immunosuppressive treatments in patients with NMOSD. Rituximab was shown to prevent relapses in NMOSD over a 72-week period in a small, randomized, double-blinded, placebo-controlled trial (
Safety and efficacy of rituximab in neuromyelitis optica spectrum disorders (RIN-1 study): a multicentre, randomised, double-blind, placebo-controlled trial.
). Since 2019, three additional disease modifying therapies (DMTs), eculizumab, inebilizumab, and satralizumab, have demonstrated high efficacy in mitigating NMOSD relapses in phase III double-blinded, randomized, placebo-controlled trials (
). Since the approval of these agents by the United States Food and Drug Administration (FDA), the cost of NMOSD treatment has increased substantially. There has never been any head-to-head comparison between the currently approved DMTs for NMOSD and the traditional immunosuppressive or anti-CD20 DMTs, despite the latter showing effectiveness in several studies (
Treatment outcomes with rituximab in 100 patients with neuromyelitis optica: influence of FCGR3A polymorphisms on the therapeutic response to rituximab.
Safety and efficacy of rituximab in neuromyelitis optica spectrum disorders (RIN-1 study): a multicentre, randomised, double-blind, placebo-controlled trial.
). Despite the availability of the newer DMTs for NMOSD in several countries, their access and affordability are still limited globally. Hence, there is still a role for the traditional therapies in the treatment of NMOSD given their established efficacy and greater ease of access.
In this study, we provide real-world evidence for the effectiveness of rituximab versus other immunosuppressive DMTs in NMOSD in a cohort of patients, the majority of whom are Black. Despite NMOSD being over-represented in Black and Asian population, phase III trials of DMTs in NMOSD have enrolled less than 12% of Black patients (eculizumab 11.9%, inebilizumab 8.6%, and satralizumab 9%) (
). Given that there is a racial disparity in disease prevalence, severity, and response to treatment in many of the autoimmune diseases, it is important to gage the therapeutic response of any DMT in a more diverse patient population.
2. Methods
2.1 Study design and patient population
This is a retrospective, observational study, for which the data were derived from archival medical records (University of Chicago NMOSD Database) from January 1, 1990, to December 1, 2020. Inclusion criteria included: (1) a diagnosis of NMOSD according to the 2015 International Panel for NMO Diagnosis (IPND) Criteria (
), (2) a positive serum IgG antibody titer for AQP4, and (3) initiation of immunosuppressant therapy within 5 years of disease onset, (4) first-line treatment with rituximab, MMF or AZA. We included results from both ELISA and cell-based assays, which were analyzed when these tests became commercially available in 2005 to confirm NMOSD diagnosis. We excluded patients with negative serum IgG antibody titers for AQP4 and those who were previously treated with immunosuppressive agents specific for multiple sclerosis (MS) (e.g., natalizumab, fingolimod, interferons). Patients were considered lost to follow up if we were unable to ascertain clinical status for at least 1 year after initiation of first-line therapy. All data were anonymized prior to analysis. The study was approved by the Institutional Review Board of the University of Chicago Medical Center, and since this was a retrospective review of data, consent was waived. This study conformed to the ethical standards of the 1964 Declaration of Helsinki.
2.2 Clinical assessment
Data collected from medical records included demographic variables, such as age, sex, race, age at onset, serologic status, date of disease onset or first attack, date of initiation of the first-line immunosuppressive treatment, end of follow-up or withdrawal of treatment, and time to first relapse after initiation of first-line immunosuppressive treatment. Relapses were defined as new CNS symptoms and signs occurring for at least 24 h and associated with new MRI lesions. Symptom worsening without corroborating MRI findings were not included in the analysis. We also documented factors associated with relapses, such as infections.
The primary outcome was time to first relapse after initiation of first-line immunosuppressive treatment. Given the limited number of patients in our cohort on oral immunosuppressive agents, we combined subjects on AZA and MMF into a single group.
2.3 Statistical analysis
Baseline characteristics were summarized using median ± range for quantitative variables, and number (percentage) for qualitative variables. Groups were compared using Wilcoxon rank-sum test or chi-square, as appropriate. Relapse-free estimates were calculated after the initiation of first-line immunosuppressive treatments by using the Kaplan-Meier method. A Cox proportional hazards regression model was used to assess the association of risk of relapse with first-line immunosuppressive treatments without and with adjustment for prespecified covariates (age at disease onset, duration of disease, sex, race, location of relapse). The proportional hazards assumption for each variable was assessed using Schoenfeld residuals.
3. Results
3.1 Demographics
A total of 167 charts were reviewed. Of these 80 were excluded on initial review as they did not meet inclusion criteria as outlined, 30 were excluded due to insufficient data, 2 received cyclophosphamide as first-line therapy, and 3 were initially misdiagnosed as MS and received MS-specific therapies as first-line treatment. Fifty-two patients were included in the study. The median age at disease onset was 34 years (range 7 – 70), and the sex ratio was 4.8:1 female to male. The median disease duration before initiation of medication was 0.5 years (range 0.02 – 5). Blacks constituted 56% of the cohort, White 23%, Hispanic 12%, and Asian 9%. Patient characteristics are summarized in Table 1.
Table 1Clinical characteristics according to first-line therapy.
Rituximab was selected as first-line therapy in 29 (55.8%) patients, while MMF was first-line therapy in 19 (36.5%), and AZA in 4 (7.7%) patients. Patients receiving MMF and AZA were grouped into a single oral immunosuppressant category. Most patients on rituximab received an induction dose of 1 g, repeated 15 days later, and then continued on maintenance therapy of 1 g every 6 months. In this real-world experience, the treatment delays between subsequent doses of rituximab were no more than 4 weeks due to scheduling and adherence issues. Furthermore, patients given rituximab shortly after their relapse were maintained on tapering doses of steroids for at least 3 weeks. B cell repletion through CD 19 cell counts was monitored systematically only in patients who had a breakthrough relapse during rituximab therapy. In 5 patients the maintenance interval of rituximab was shortened to every 3–4 months due to early CD19 repletion. Most patients treated with MMF received 1 g twice a day (79%), while the remaining (21%) received either 500 mg or 750 mg twice per day. Azathioprine dose was more variable, ranging from 1 to 3 mg/kg/d.
3.2 Effectiveness
A total of 7 of 29 patients (24%) receiving rituximab experienced a relapse within the first three years of treatment, compared to 13 of 23 (57%) patients receiving either AZA or MMF. Within the first 6 months of treatment, 2 (6.9%) patients treated with rituximab experienced a relapse, compared to 7 (30.4%) patients treated with either MMF or AZA. In the 29 patients treated with rituximab, the 1-year and 3-year proportion of relapse-free patients was 88.8% and 70.9%. For the 23 patients treated with either AZA or MMF, the 1-year and 3-year proportion of relapse-free patients was 69.5% and 38.7%. In the univariate analysis, the risk of relapse was significantly higher in patients treated with AZA or MMF compared to those treated with rituximab (hazard ratio [HR] of 2.48 [0.99 – 6.21]; p = 0.046) (Fig. 1). In the multivariate analysis, adjustments were made for age at disease onset, race, sex, duration of disease, location of lesions, presence of comorbidities, and other autoimmune diseases. The Black cohort of patients exhibited a significantly lower risk of relapse compared to White (HR = 0.17 [0.03 – 0.83]; p = 0.029). Other covariates included in the multivariate model did not show a significant association with risk of relapse (Table 2).
Fig. 1Kaplan-Meier curve – proportion of relapse-free patients following initiation of first line immunosuppressive therapy.
Three patients (10.3%) treated with rituximab discontinued therapy during the study period due to insurance issues, and one (3.4%) transitioned to oral immunosuppressants due to insurance coverage. Eight patients (34.8%) treated with oral immunosuppressants were eventually transitioned to rituximab therapy, and three patients (13%) to new monoclonal antibody therapy within fifteen years of oral immunosuppressant initiation. Two patients (8.7%) received combination therapy within five years of immunosuppressant initiation (MMF and rituximab). These patients were included in the analysis as intention to treat. Rituximab infusions were generally well tolerated without any discontinuation due to adverse events or clinically significant hypogammaglobulinemia. Over the observation period, 1 patient on rituximab was hospitalized for symptomatic worsening due to urinary tract infection (UTI) but without MRI worsening. Otherwise, UTIs were infrequently observed (6 in the rituximab and 4 in the oral treatment cohorts), mild in severity, and were managed as outpatient.
4. Discussion
The results of this study demonstrate that the risk of relapse for patients treated with MMF or AZA was significantly higher than for patients treated with rituximab, with a HR of 2.48 [0.99 – 6.21]; p = 0.046. The 1-year and 3-year relapse-free proportions were also significantly higher in patients receiving rituximab compared to those receiving either AZA or MMF. Black patients exhibited a significantly lower risk of relapse compared to Whites, with a HR of 0.17 (0.03 – 0.83); p = 0.029. The mechanism for this difference in outcomes is not clearly understood but could be in part due to early diagnosis because of clinical presentation of a more fulminant disease in the Black cohort (
), and hence rapid initiation of DMTs and aggressive treatment of relapses, at least in our cohort. Also, the White cohort was underrepresented in this study, which could potentially bias the results towards the Black cohort. There were otherwise no significant associations in the multivariate analysis, adjusting for covariates, including duration of disease and location of lesions.
The results of this study are in line with previous observations.
reported a significantly higher relapse risk in patients treated with MMF compared to those treated with rituximab (HR= 1.66 [1.03–40.8]; p = 0.042). This difference was not evident when compared to patients treated with AZA (HR= 2.22 [0.61–4.5]; p = 0.32). Conversely,
found similar relapse-free proportions in rituximab-treated patients compared to MMF-treated patients (HR= 1.48 [0.75–2.93]; p = 0.26), but risk of relapse was significantly higher in AZA-treated patients (HR= 2.12 [1.12–4.01]; p = 0.002).
reported a higher relapse risk in patients receiving AZA compared to those treated with rituximab (HR= 1.82 [1.1–3.1]; p = 0.03), but no statistically significant difference between MMF versus rituximab (HR= 1.27 [0.7–2.5]; p = 0.48). Heterogeneity in the sample size and population characteristics, as well as study design and methodology could account for some differences. Nonetheless, all three studies reported a significant reduction in relapse rate after initiation of any one of these immunosuppressants.
The differential treatment effect between rituximab and the oral immunosuppressants is not clearly understood. Anti-AQP4 antibodies exert their pathologic effect through complement mediated destruction of astrocytes and secondary injury to oligodendrocytes and myelin (
Pathogenesis of autoimmune demyelination: from multiple sclerosis to neuromyelitis optica spectrum disorders and myelin oligodendrocyte glycoprotein antibody-associated disease.
). During a NMOSD relapse, there is activation, proliferation and cytokine secretion by both T and B cells. T cells, particularly Th17 cells, mediate the endothelial injury in NMOSD inflammation, leading to recruitment of other cells types that further exacerbate tissue injury, such as neutrophils and eosinophils (
The expansion of circulating IL-6 and IL-17-secreting follicular helper T cells is associated with neurological disabilities in neuromyelitis optica spectrum disorders.
). Rituximab is a chimeric monoclonal antibody against CD20, a transmembrane protein expressed in all B cells. After binding to its antigen, rituximab triggers a complement-dependent cytotoxic effect, leading to depletion of early pre-B, mature, and memory B cells (
). B cells and their proinflammatory cytokines, plasmablast derivatives, and diminished regulatory function all play a central role in NMOSD pathology (
). Through these mechanisms, these agents are all capable of ameliorating the inflammatory cascade in NMOSD. In addition to the variability in the mechanisms of these DMTs, the time of onset of maximal effects on the immune cells (the pharmacodynamic effects) is quite variable among these agents, with delayed effects of oral DMTs. Dose escalation of therapy initially could further delay the achievement of full pharmacodynamic effect. Hence, studies that examine time to first relapse as an outcome measure using different DMTs with dissimilar mechanism of action should be interpreted with caution and take into account the time needed to achieve full drug effect. This is particularly important when comparing across different trials involving DMTs with distinctive mechanisms of action. It may be judicious to evaluate the effectiveness of a particular DMT after a certain “run-in” period, when the full effects of that DMT have been achieved.
The key finding of our study is that the proportion of relapses in the first 6 months of treatment was significantly different in patients receiving rituximab compared to those receiving oral immunosuppressants. In the rituximab group, 2 (6.9%) patients experienced a relapse within 6 months of starting therapy despite B cell count of 0, compared to 7 (30.4%) patients in the second group. This likely reflects the longer time to onset of clinical effect of MMF or AZA compared to rituximab. Prospective and retrospective studies analyzing the use of either AZA or MMF in other autoimmune diseases such as myasthenia gravis (MG), systemic lupus erythematosus (SLE), and inflammatory bowel disease (IBD) suggest that the therapeutic effects of these medications can be delayed by as much as 1 year (
Lack of effect of intravenous administration on time to respond to azathioprine for steroid-treated Crohn's disease. North American azathioprine study group.
). Rituximab, on the other hand, is believed to exert a more rapid effect on B cells, with depletion of circulating cells within 24 h of administration (
Despite a more rapid pharmacodynamic effects of rituximab on B cell depletion, its full clinical effectiveness in some NMOSD patients may take several weeks to be fully achieved (
). The mechanism for this phenomenon is not clearly understood, but could be due to inadequate B cell depletion, especially in the lymphoid tissue, highly active NMOSD disease in certain susceptible individuals, or paradoxical, transient, increase in proinflammatory cytokines induced by rituximab itself (
). This rebound of inflammatory activity after rituximab administration is thought to be due to transient elevation of tumor necrosis factor alpha (TNF-α), IL-6, and B-cell activating factor (BAFF), the latter 2 cytokines known for their role as growth factors for plasmablasts (
). Given that the time shortly after an acute NMOSD relapse may be a period of heightened sensitivity for an exacerbation of ongoing inflammatory activity, which could be triggered by a variety of sources, including paradoxically by rituximab, prolonged steroid taper (> 1 month) overlapping with rituximab has been advocated as a strategy (
Safety and efficacy of rituximab in neuromyelitis optica spectrum disorders (RIN-1 study): a multicentre, randomised, double-blind, placebo-controlled trial.
(RIN-1), the prednisolone dosage was fixed for 2 months (5–30 mg) following initiation of rituximab and subsequently tapered slowly by 10% (1–3 mg) every 4 weeks, with no relapses occurring in the rituximab treated arm. In
), prednisone overlapped with inebilizumab for only 21 days following initiation, and the majority of relapses in that study occurred early on, suggesting that longer overlap of corticosteroids (as in the RIN-1 trial) may be necessary to prevent early relapses after initiation of B-cell depleting agents. For starting oral immunosuppressive treatments, a prolonged oral steroid taper schedule may be needed to allow the pharmacodynamic effects of these agents to be fully achieved, 3 months for MMF and 6 months for AZA (
Our study adds to the current body of evidence demonstrating the effectiveness of rituximab and oral immunosuppressants in the management of NMOSD relapses, each class of treatments associated with unique nuances of therapeutic efficacy, mode of action, and pharmacodynamic profile. Furthermore, our study along with the recently published observations by
, taken together, demonstrate effectiveness of rituximab in a large cohort of NMOSD patients, majority of whom are Black (51–56%), a subgroup that has been historically underrepresented in the phase III trials of DMTs for NMOSD. The expanded racial composition of both studies increases its generalizability on a global scale, a significant consideration in rare diseases that are more rigorously evaluated when non-White subjects are sufficiently represented. This study also used time to first relapse and relapse-free proportion as outcome measures, to better correlate with NMOSD phase III trials.
The main limitations of our study are the retrospective design and dosing variability of agents used. To make our cohort as homogenous as possible, we included only anti-AQP4 antibody positive patients and rigorously collected all data timepoints to ensure completeness of our database. In addition, we included only MRI-confirmed relapses and excluded those patients in whom relapses were triggered by active infections or recent pregnancy. Although baseline characteristics between the two groups were similar overall, the groups differed in two aspects: (1) duration of disease, and (2) proportion of spinal lesions, which were significantly higher in the oral immunosuppressant group. Differences in baseline characteristics could introduce bias into study results, as it is possible that patients receiving oral immunosuppressants had a more severe disease course, which could account for some of the treatment differences. In addition, we did not collect baseline EDSS in our population. Low numbers in the AZA-treated patients prompted us to group MMF and AZA, which can also lead to dilution of the true effect of either medication.
5. Conclusion
In this real-world, urban cohort of seropositive NMOSD patients, the majority of whom are Black, rituximab was more effective in preventing relapses compared to AZA and MMF, with a 1-year and 3-year proportion of relapse-free patients at 88.8% and 70.9%.
Author's contributions
Laura Dresser and Adil Javed had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of data analysis.
Study concept and design: Laura Dresser and Adil Javed.
Acquisition of data: Laura Dresser, Adil Javed
Drafting of the manuscript: Laura Dresser, Adil Javed
Critical revision of the manuscript for intellectual content: Laura Dresser, Widad Abou Chaar, Amanda Frisosky Abuaf, Anthony T. Reder, Veronica P. Cipriani, Adil Javed
Statistical analysis: Laura Dresser
Study supervision: Adil Javed
Funding source
The authors did not receive any specific financial support for the research or authorship of this article.
Declaration of Competing Interest
AJ has received honoraria and consulting fees from Biogen, Serono, Genentech-Roche, BMS, and TG therapeutics.
VPC has received compensation for consulting and/or speaking engagements from EMD Serono, Sanofi and Genentech Roche.
ATR has received unrestricted grant support from Roche/Genentech.
AFA has received fellowship funding from Roche/Genentech.
LD and WAC have nothing to disclose.
Acknowledgements
The authors thank the patients who contributed to this study.
References
Appel G.B.
Contreras G.
Dooley M.A.
Ginzler E.M.
Isenberg D.
Jayne D.
Li L.-S.
Mysler E.
Sánchez-Guerrero J.
Solomons N.
et al.
Mycophenolate mofetil versus cyclophosphamide for induction treatment of lupus nephritis.
Treatment outcomes with rituximab in 100 patients with neuromyelitis optica: influence of FCGR3A polymorphisms on the therapeutic response to rituximab.
Pathogenesis of autoimmune demyelination: from multiple sclerosis to neuromyelitis optica spectrum disorders and myelin oligodendrocyte glycoprotein antibody-associated disease.
The expansion of circulating IL-6 and IL-17-secreting follicular helper T cells is associated with neurological disabilities in neuromyelitis optica spectrum disorders.
Lack of effect of intravenous administration on time to respond to azathioprine for steroid-treated Crohn's disease. North American azathioprine study group.
Safety and efficacy of rituximab in neuromyelitis optica spectrum disorders (RIN-1 study): a multicentre, randomised, double-blind, placebo-controlled trial.
00222-5&id=gr1.jpg){kind=link}