Exploring rituximab for the treatment of refractory myasthenia gravis: a single-centre experience

Abstract

Background

Myasthenia gravis (MG) is a rare autoimmune disorder that affects neuromuscular transmission, leading to muscle weakness. While most patients respond to standard treatments, approximately 15% remain refractory, prompting interest in alternative therapies. This study examined the use of rituximab in patients with refractory MG at a single centre in Latvia, intending to broaden discussions on treatment strategies.

Materials and methods

The prospective cohort study was conducted at Pauls Stradiņš Clinical University Hospital from November 2022 to March 2024. Refractory MG was defined as failure to achieve adequate disease control despite standard immunosuppressive therapy, including patients with persistent symptoms, recurrent crises or intolerance to medications. Myasthenia Gravis Composite Score (MGCS), Myasthenia Gravis Activities of Daily Living and Myasthenia Gravis Quality of Life (scores at baseline and monthly for 6 months) were completed. Glucocorticosteroid doses were also recorded. Data were analysed using descriptive statistics and Wilcoxon signed-rank test, considering p<0.05 significant.

Results

98 patients with MG were evaluated to identify treatment-refractory cases. A total of nine patients with refractory MG (median disease duration 7 years, range 1–12) were included and received a single low dose of rituximab. Among the nine patients treated, seven were acetylcholine receptor-positive and two were muscle-specific kinase antibody-seropositive. 6 months after treatment, the median MGCS decreased from 11.5 (IQR 7.3–17.8) to 7.5 (IQR 3.8–8.0) (p=0.025). The median daily corticosteroid dose decreased from 15.0 mg (IQR 10.0–20.9) to 8.5 mg (IQR 5.0–12.5) (p=0.036). There were no exacerbations of MG or significant adverse reactions.

Conclusions

Our findings suggest that rituximab may be a promising treatment option for refractory MG. The notable clinical benefits and safety profile make rituximab a valuable addition to this challenging autoimmune disorder treatment options. Further research is needed to refine patient selection and dosing for optimal treatment outcomes.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Over the past 25 years, rituximab has shown promising effectiveness as an adjunctive treatment, especially in refractory cases of myasthenia gravis. However, key questions still exist about optimal dosing, retreatment schedules and its effectiveness in acetylcholine receptor (AChR)-positive subtypes.

WHAT THIS STUDY ADDS

• In this study, we present a small group mainly consisting of AChR-positive patients, with outcomes that match previously reported positive responses. By confirming existing findings, our data add to the growing amount of real-world evidence supporting rituximab’s therapeutic role in this rare autoimmune disease.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• These observations provide practical insights that could help guide future treatment protocols and support considerations for regulatory label expansion of this well-established, but currently off-label, therapy.

Introduction

Myasthenia gravis (MG) is a rare autoimmune neuromuscular disorder characterised by impaired synaptic transmission at the neuromuscular junction, causing fluctuating weakness in different muscle groups. Immunologically, most patients with MG are acetylcholine receptor antibody seropositive (AChR+), while approximately 10% of patients are muscle-specific kinase antibody seropositive (MuSK+). Furthermore, some patients exhibit undetectable antibody levels and are classified as seronegative. Although MG is a rare disease, it significantly impacts the health of >1 million people worldwide, with an estimated prevalence of 150–250 cases per million and an annual incidence of 8–10 cases per million.^{1 2} In Latvia, data from 1 January 2015 indicated that the prevalence rate of MG was 113.8 per million and the incidence rate was 9.7 per million per year.³

Although most patients with MG respond favourably to standard treatments, including acetylcholinesterase inhibitors, corticosteroids, intravenous Ig (IVIg), plasma exchange (PEX) and corticosteroid-sparing immunosuppressants, a substantial subset, approximately 15%, exhibits refractory MG.⁴ These patients often require prolonged treatment with high-dose glucocorticoids, which are associated with severe adverse effects. Patients with refractory MG have a reduced quality of life and sometimes require hospitalisation for potentially lethal exacerbations; due to these challenges, there is growing interest in exploring alternative therapies for MG.⁵

Rituximab, a monoclonal, chimeric anti-CD20 antibody, has emerged as a promising option for refractory MG due to its unique ability to deplete B cells through antibody-dependent cytotoxicity, complement-dependent cytotoxicity and apoptosis.⁶ A single dose of this anti‐CD20 monoclonal antibody induces nearly complete B-cell depletion in the peripheral blood.⁷ This mechanism not only facilitates significant clinical improvements in patients with MG but also exerts a steroid-sparing effect, particularly in MuSK+ patients. In recent decades, rituximab has been used in AChR+, MuSK+ and seronegative MG patients, ranging from newly diagnosed to refractory to conventional treatments.^{8 9} Evidence from retrospective studies and meta-analyses suggests that adding rituximab to the standard treatment regimen results in a substantial clinical improvement in both AChR+ and MuSK+ patients.^{10 11} In a recent randomised controlled trial in Sweden, a single dose of rituximab was associated with a greater probability of minimal MG manifestations and reduced need for rescue medications in AChR+ patients with new-onset MG.¹² Although effective, the use of rituximab had remained off-label until a recent position paper by five Nordic countries recommended its inclusion as an adjunct to first-line therapy for newly diagnosed or refractory cases of generalised MG. This study provided a detailed exploration of the potential of rituximab as a transformative treatment for patients with resistant MG.¹³ The authors suggested that this approach could reduce exacerbations and hospitalisations, thus improving quality of life.

Drawing inspiration from the work of Nordic countries, we aimed to share our single-centre experience from Latvia. We evaluated the results of administering a single low dose of rituximab to patients with refractory MG. We hope to provide additional information and contribute to broader discussions on optimising treatment strategies for this complex disease.

Materials and methods

A prospective cohort study was conducted in the Department of Neurology at Pauls Stradiņš Clinical University Hospital from November 2022 to March 2024. This was an exploratory pilot study based on a small, real-world cohort of patients with refractory MG treated at a single tertiary centre. Given the rarity of this condition and the constrained timeframe, no formal power calculation was performed. All subgroup analyses (eg, AChR+ vs MuSK+) were exploratory in nature and not powered to detect statistically significant differences.

This study enrolled patients diagnosed with generalised refractory MG, classified as refractory according to at least one of the following criteria:

1. Recurrent admissions with a myasthenic crisis requiring rescue treatment such as IVIg or PEX procedures.

2. High corticosteroid doses (>20 mg/daily) used and inability to lower the dose without deterioration of symptoms.

3. Severe side effects from corticosteroids and immunosuppressants, suboptimal MG compensation without immunosuppression.

4. Suboptimal disease control, defined as a Myasthenia Gravis Composite Score (MGCS) ≥10 and/or Myasthenia Gravis Activities of Daily Living (MG-ADL) score ≥6, despite adequate immunosuppressive therapy.

Before rituximab, all patients had been treated with corticosteroids and at least one immunosuppressive agent (azathioprine, mycophenolate mofetil or methotrexate). Three patients received IVIG for myasthenic crises, but no patients were treated with FcRn antagonists, complement inhibitors or maintenance PLEX. All patients received a single intravenous dose of rituximab: eight received 500 mg and one with a body mass index >35 kg/m² received 1000 mg. To ensure tolerance to the infusion, all patients were premedicated with intravenous methylprednisolone (125 mg), chloropyramine hydrochloride (20 mg) and paracetamol (1000 mg).

The neurological status of each patient was evaluated with the MGCS, MG-ADL and Myasthenia Gravis Quality of Life (MG-QoL) scores, which were evaluated at the beginning of the study and monthly for 6 months after rituximab infusion. Evaluations were conducted during scheduled monthly follow-up visits at the Outpatient Clinic of Pauls Stradins Clinical University Hospital. All assessments were administered by trained clinical staff and provided in the patients’ native language to ensure full comprehension. MGCS is a physician-administered assessment and includes 10 items evaluating objective clinical symptoms and their severity, covering muscle strength in arms, legs, facial muscles, respiratory and bulbar function, and response to fatiguability tests. Total scores range from 0 to 50, with higher scores indicating greater impairment. A change of ≥3 points is considered clinically meaningful.¹⁴ MG-ADL and MG-QoL are patient-reported evaluations of the ability to perform daily tasks and perception of health-related quality of life (associated particularly with myasthenia). In both scales, highest score indicates greater disability and impairment. A change of ≥3 points was considered clinically significant for MG-QoL and ≥2 points for MG-ADL.^{15 16} Corticosteroid doses were also recorded at each visit. During the 6-month follow-up, no changes were made to concomitant immunosuppressive medications. Patients continued their baseline doses of azathioprine or mycophenolate mofetil throughout the observation period.

Demographic data from the patients were analysed using descriptive statistics. Categorical data were presented as counts and percentages, and continuous data were presented as mean and SD for normally distributed data. Given the small sample size (n=9) and limited power of normality tests, all continuous variables were analysed using the non-parametric Wilcoxon signed-rank test, and data are presented as medians (IQR). Data analysis was performed with Jamovi (V.2.3). A p value <0.05 was considered to indicate a statistically significant difference. Also, the p values are descriptive; no multiplicity adjustment applied due to exploratory design.

Patient and public involvement statement

No ‘patient and public involvement’. The patients were not directly involved in the design, conduct, reporting or dissemination plans of this investigation. However, the results of the study were discussed with the patients during routine clinical follow-up, and their feedback on the tolerability of the treatment was taken into account in the clinical evaluation.

Results

We monitored 98 patients with MG at Pauls Stradiņš Clinical University Hospital from November 2022 to March 2024. Within this cohort, nine patients (9.1%) were classified as having refractory MG according to the established criteria and were treated with a single low dose of rituximab. This subgroup included five women and four men (median age: 47 years, IQR 36–62), where the majority of patients had late-onset myasthenia gravis (cut-off age was 50). While thymectomy is recommended for AChR+ early-onset myasthenia gravis (EOMG) patients, only two had undergone surgery before rituximab. The remaining patients either had no radiological evidence of thymoma or were not referred due to age or comorbidities at diagnosis. Five patients had experienced severe complications from extended steroid therapy, including recurrent infections, weight gain and osteoporosis. The median daily dose of prednisone before rituximab treatment was 15.0 mg (IQR 10.0–20.9). The detailed demographic and clinical characteristics of these patients are summarised in table 1.

Table 1. Baseline characteristics of patients with refractory MG.

ID	Age of onset	Duration of MG (years)	Antibodies	Baseline therapy	Myasthenic crises (n)	MGCS	MG-ADL	MG-QoL
1	EOMG	7	AChR	Pyridostigmine Prednisone (15 mg) Azathioprine	3	4	0	21
2	EOMG	7	AChR	Pyridostigmine Prednisone (27.5 mg) Azathioprine	4	17	12	35
3	LOMG	1	MuSK	Prednisone (30 mg) Mycophenolate mofetil	1	18	7	38
4	EOMG	12	AChR	Pyridostigmine Prednisone (10 mg) Mycophenolate mofetil	0	21	9	43
5	EOMG	7	MuSK	Prednisone (5 mg) Mycophenolate mofetil	1	19	10	24
6	EOMG	4	AChR	Pyridostigmine Prednisone (5 mg) Mycophenolate mofetil	2	7	2	48
7	EOMG	2	AChR	Pyridostigmine Prednisone (22.5 mg) Azathioprine	1	6	5	16
8	LOMG	3	AChR	Pyridostigmine Methylprednisone (16 mg) Azathioprine	1	11	6	37
9	LOMG	7	AChR	Pyridostigmine Prednisone (10 mg) Azathioprine	2	8	6	10

Suboptimal disease control is defined as MGCS≥10 or MG-ADL≥6 in the absence of crisis.

AChR, acetylcholine receptor; EOMG, early-onset myasthenia gravis (<50 years old); ID, identification number; LOMG, late-onset myasthenia gravis; MG, myasthenia gravis; MG-ADL, Myasthenia Gravis Activities of Daily Living; MGCS, Myasthenia Gravis Composite Score; MG-QoL, Myasthenia Gravis Quality of Life; MusK, muscle-specific kinase.

6 months after initiation of rituximab therapy, median MGCS decreased from 11.5 (IQR 7.3–17.8) to 7.5 (IQR 3.8–8.0), indicating a statistically significant improvement (*p=0.025, Wilcoxon signed-rank test), though AChR+ patients showed heterogeneous responses. In particular, MuSK+ patients exhibited a dramatic decrease in the MGCS, from 18.5 to 2, consistent with previous reports of greater responsiveness in this subgroup. In contrast, the reduction in the MGCS for AChR+ patients was more modest, from 10.6 to 7.1, and this change did not achieve statistical significance (p=0.072). The optimal response among AChR+ patients was observed 3 months after rituximab infusion, followed by a slight increase in scores at 6 months. One AChR+ patient showed mild deterioration relative to baseline (figure 1).

Figure 1. Myasthenia Gravis Composite Score (MGCS) changes post-rituximab (RTX). *p=0.025 at 6 months vs baseline. Note: Patient 7 (acetylcholine receptor-positive) showed paradoxical worsening. ID, identification number.

Similar improvements were observed in the MG-ADL score, decreasing from 6.0 (IQR 3.0–8.0) to 3.0 (IQR 1.0–4.0) at 6 months (p=0.096). Both MuSK+ patients achieved near-complete symptom resolution (MG-ADL=0), consistent with established literature. Both MuSK+ patients achieved near-complete symptom resolution (MG-ADL=0), consistent with established literature (figure 2).

Figure 2. The Myasthenia Gravis Activities of Daily Living (MG-ADL) score each month after rituximab (RTX) infusion. ID, identification number.

Although the MG-QoL score improved from 30 (IQR 22–42) to 23 (IQR 15–33), 6 months after rituximab infusion, this change did not reach statistical significance (p=0.173), reflecting inter-individual variability in self-perceived outcomes.

At 6-month follow-up, six out of nine patients (66.7%) achieved corticosteroid dose reduction, while three maintained their baseline dose (figure 3). One AChR+ patient discontinued steroids entirely. The median daily corticosteroid dose decreased from 15.0 mg (IQR 10.0–20.9) to 8.5 mg (IQR 5.0–12.5) (*p=0.036, Wilcoxon signed-rank test). No adjustments were made to baseline immunosuppressant regimens; all patients remained on their pretreatment doses of azathioprine or mycophenolate.

Figure 3. Monthly variations in corticosteroid doses after rituximab infusion. ID, identification number.

Within 6 months after rituximab infusion, no patients experienced MG exacerbations that required IVIg or PEX. Ig levels remained stable, with no cases of hypogammaglobulinaemia. The median IgG level was 9.3 g/L (IQR 7.1–11.5) before treatment and 9.7 g/L (IQR 7.3–12.1) at 6 months, while the median IgM level was 0.9 g/L (IQR 0.6–1.1) before treatment and 1.0 g/L (IQR 0.7–1.2) post-treatment. Furthermore, there were no documented adverse reactions during rituximab infusions, nor were there cases of cytopenia or opportunistic infections after infusion. All adverse events were assessed using the Common Terminology Criteria for Adverse Events (CTCAE V.5.0). One patient experienced a self-limited influenza-like illness 4 months after infusion (grade 1 CTCAE), requiring no medical intervention. No grade≥2 events were observed, and there were no infusion reactions, opportunistic infections or cases of cytopenia.

CD19+ B-cell counts were available for three patients. B-cell monitoring at 6 months showed depletion (median 2 cells/µL, range: 2–25) in three tested patients; others lacked data due to clinical constraints. Although CD19+ B-cell counts were not available for every patient, the data from these three individuals confirmed effective peripheral B-cell depletion consistent with rituximab’s mechanism of action.

Given the small sample size, particularly in the MuSK+ group (n=2), statistical interpretation is limited, and findings should be considered descriptive.

Discussion

The results of our single-centre study on the use of rituximab for refractory MG offer promising evidence of its efficacy and safety. 6 months after rituximab administration, our patients with MG exhibited improvements in all clinical scores, with no adverse effects. It is important to note that, although one patient experienced a slight increase in the MGCS 6 months after treatment, the causality cannot be definitively established; the episode was self-limited and did not require hospitalisation or specific antiviral therapy. No signs of severe immunosuppression, such as hypogammaglobulinaemia or opportunistic infection, were observed in this patient. Importantly, this patient showed clinical improvement based on the MG-ADL score, which underlines the overall positive impact of treatment.

Rituximab demonstrated notable efficacy in MuSK+ patients, resulting in a significant improvement in the MGCS and the MG-ADL score, nearly reaching zero. This finding aligns with previous research, suggesting that MuSK+ patients may derive greater benefit from B-cell-depleting therapies.¹⁷ The results for the AChR+ group were less favourable than the MuSK+ group. The observed difference in treatment response between MuSK+ and AChR+ patients may be partially explained by known immunological mechanisms, as described in previous studies.

Rituximab depletes CD20-expressing pre-B and mature B cells, but does not directly affect plasma cells, which lack CD20 expression. Its effect is therefore indirect, reducing the generation of new antibody-secreting cells. The more robust response in MuSK+ patients may reflect greater dependency on B-cell-driven IgG4 autoantibody production, which may be more sensitive to upstream B-cell depletion.^{18 19} However, in AChR+ MG, the main antibody subclasses are IgG1 and IgG3 from long-lived plasma cells generated by a T-helper 1 response. These cells, once established in the bone marrow, are less affected by rituximab, resulting in reduced effectiveness for this subtype. The variation in plasma cell longevity and immune pathways is key to understanding the differential effects of rituximab in MG.²⁰ While mechanistic differences may contribute (per references16,18), this remains speculative without B-cell/antibody subtyping data. Although the effectiveness of this monoclonal antibody in treating AChR+ MG remains debated,^{17 21 22} our data indicate that, despite variable response and sustained benefit in only 2/7 cases, in general, AChR+ patients improved clinically and did not require rescue treatment for 6 months after receiving the infusion. This finding is in line with the previous randomised controlled trial on rituximab in AChR+ MG and suggests that rituximab could provide a valuable adjunct to individualised treatment protocols.^{12 23} However, our study did not directly assess B-cell or antibody subtypes; thus, these interpretations remain speculative.

In our cohort, clinical responses among AChR+ patients were variable. While some showed transient improvement in MGCS and steroid reduction, only two demonstrated sustained benefit at 6 months. These findings align with previous literature suggesting that AChR+ MG may respond less predictably to rituximab, possibly due to a greater role of long-lived plasma cells resistant to CD20-targeted therapies. Among AChR+ patients, those with disease duration <5 years demonstrated greater clinical improvement and steroid tapering, suggesting that earlier initiation of rituximab may be associated with better outcomes. Our observation that shorter-disease-duration AChR+ patients responded better aligns with Piehl et al’s trial in new-onset MG, though their 1000 mg dose contrasts with our 500 mg regimen.¹² However, our sample size precludes definitive conclusions.

Further, rituximab therapy induced significant steroid-sparing effects. This steroid dose reduction is particularly crucial for patients with MG, who are typically subjected to prolonged corticosteroid treatment and their associated severe complications, such as recurrent infections, drug-resistant hypertension, fluid retention, weight gain and insomnia. In our study, two-thirds of the patients (66.7%) successfully decreased their steroid dose while achieving clinical improvement. This finding is consistent with the literature; for example, Nowak et al²⁴ reported a mean reduction in the prednisone dose of 65.1%, 85.7% and 93.8% after the first, second and third cycles of rituximab therapy, respectively, highlighting rituximab’s efficacy in managing steroid dependence across different studies, enhancing both quality of life and potentially reducing the morbidity associated with treating refractory MG.

Despite clinical improvements in patients with refractory MG following rituximab therapy, we did not observe statistically significant changes in the MG-QoL score. While the overall average MG-QoL score decreased slightly, one patient experienced an increase in the score despite clinical improvement. The non-significant change in MG-QoL may reflect the complexity of patient experiences and psychological burden in MG, which are not always paralleled by physical improvements. These results align with findings from a large German cohort study, which emphasised that, despite medical advancements, many patients with MG continue to experience significant impacts on their quality of life, particularly in terms of mental well-being.²⁵ Additionally, a single-centre cross-sectional study conducted by colleagues in Poland underscores that, although factors such as MG severity, age of onset and employment status significantly impact the quality of life, clinical improvements resulting from treatments do not consistently translate into statistically significant enhancements in the MG-QoL score.²⁶ This highlights the complex and multifaceted nature of MG and its management. In our study, the persistence of quality-of-life issues despite the clinical improvement could be attributed to factors such as prolonged steroid use before rituximab treatment, the severe course of the disease and a lack of psychological support and rehabilitation for this patient group. These areas are crucial and should be addressed by future research to enhance overall patient well-being.

Of particular note, there were no severe side effects, such as cytopenia or opportunistic infections, after rituximab infusion. Furthermore, the IgG levels remained consistent, suggesting no occurrence of hypogammaglobulinaemia, a typical concern with B-cell-depleting therapies such as rituximab. This finding aligns with other studies that also affirm the long-term safety and efficacy of rituximab for the treatment of MG. These studies collectively report minimal significant adverse effects and stable IgG levels, mitigating concerns about common complications of B-cell depletion.^{27 28} This favourable safety profile underscores the potential of rituximab as a sustainable option for the long-term management of refractory MG.

While preliminary trends are observed—particularly in the MuSK+ group—these findings must be interpreted cautiously. The primary aim of this work is to contribute real-world observational data and generate hypotheses for larger, controlled studies.

Limitations

Despite the encouraging results, the small, single-centre cohort from a tertiary referral hospital and lack of power calculations limit the generalisability of the findings. The study was not powered to detect inter-subgroup differences; accordingly, the findings should be regarded as exploratory and hypothesis-generating. This limitation is particularly relevant to the MuSK-positive subgroup, in which the small sample size precludes reliable statistical inference. Another limitation of this study is the short follow-up period, which prevents evaluation of the long-term durability of rituximab response or the timing of potential re-dosing.

Conclusions

Our experience reinforces the role of rituximab as a valuable treatment alternative for patients with refractory MG, particularly those with MuSK antibodies but also AChR+ patients. Given its significant clinical benefits and manageable safety profile, rituximab presents a viable option in the therapeutic arsenal against this challenging autoimmune disorder. More studies are needed to refine patient selection and optimise dosing regimens to maximise outcomes and minimise risks.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.