Article Text

Rituximab in AChR subtype of myasthenia gravis: systematic review
  1. Vincenzo Di Stefano1,2,
  2. Antonino Lupica3,
  3. Marianna Gabriella Rispoli2,
  4. Antonio Di Muzio4,
  5. Filippo Brighina1,
  6. Carmelo Rodolico3
  1. 1 Department of Biomedicine, Neuroscience and advanced Diagnostic, University of Palermo, Palermo, Sicily, Italy
  2. 2 Department of Neuroscience Imaging and Clinical Sciences, Gabriele d'Annunzio University of Chieti and Pescara, Chieti, Abruzzo, Italy
  3. 3 Department of Clinical and Experimental Medicine, Unit of Neurology and Neuromuscular Disease, University of Messina, Messina, Sicilia, Italy
  4. 4 Department of Neurology, SS Annunziata Hospital, Chieti, Abruzzo, Italy
  1. Correspondence to Dr Vincenzo Di Stefano, Department of Biomedicine, Neuroscience and advanced Diagnostic, University of Palermo, Palermo, Sicily, Italy; vincenzo19689{at}


Myasthenia gravis (MG) is a chronic autoimmune disorder of the neuromuscular junction characterised by an autoantibody against acetylcholine receptor (AChR-Ab), autoantibody against muscle-specific kinase (MuSK-Ab), lipoprotein-related protein 4 or agrin in the postsynaptic membrane at the neuromuscular junction. Many patients are resistant to conventional treatment and effective therapies are needed. Rituximab (RTX) is a monoclonal antibody directed against CD20 antigen on B cells which has been successfully employed in anti-MuSK-Ab+MG, but the efficacy in anti-AChR-Ab+MG is still debated. The purpose of this systematic review was to describe the best evidence for RTX in the acetylcholine receptor subtype. The authors undertook a literature search during the period of 1999–2019 according to the Preferred Reporting Items for Systematic Reviews and Meta-Analys methodology, employing (myasthenia)+(gravis)+(RTX) as search terms. The analysis was confined to studies that include at least five patients with confirmed anti-AChR-Ab+MG. Thirteen studies have been selected, showing a good safety. The data obtained were heterogeneous in terms of posology, administration scheme and patients’ evaluation, ranging from a minimum of two to a maximum of three cycles. RTX led to a sustained clinical improvement with prolonged time to relapse, in parallel to a reduction or discontinuation of other immunosuppressive therapies. Treatment with RTX appears to work in some but not all patients with anti-AChR-Ab+MG, but randomised controlled trials are needed. Future studies should take into account the subtype of MG and employ reliable measures of outcome and severity focusing on how to identify patients who may benefit from the treatment. Trial registration number: NCT02110706.

  • myasthenia
  • immunology
  • neuromuscular
  • neuroimmunology

Statistics from


Myasthenia gravis (MG) is an antibody-mediated chronic autoimmune disorder of the neuromuscular junction characterised by skeletal muscle weakness and fatigability with an estimated annual incidence of 1–2 per 100 000 and a prevalence of 20–50 per 100 000.1 2 The majority of patients with MG present with an autoantibody against acetylcholine receptor (AChR-Ab), whereas the remaining show autoantibody against muscle-specific kinase (MuSK-Ab), lipoprotein-related protein 4 or agrin in the postsynaptic membrane at the neuromuscular junction.3

Nevertheless, in some cases, a specific antibody profile cannot be identified, but cell-based assays were recently established to improve the detection of antibodies in patients who were previously seronegative.4 5

Although many immunosuppressive and immunomodulatory therapies are available, including steroids, azathioprine, mycophenolate mofetil, methotrexate and cyclosporine, up to 20% of patients with MG are refractory to conventional treatment, showing incomplete responses to multiple combined therapies or requiring repeated immunoglobulin (IVIg) infusions or plasma exchange.6 7 Refractory MG decreases the quality of life of affected patients, and sometimes, it requires hospitalisation for potentially lethal exacerbations.8 So, it is relevant to find out alternative effective treatments to improve the prognosis in this condition. On this perspective, new promising monoclonal antibodies (ie, eculizumab, rituximab (RTX) and ocrelizumab) have appeared in the last decade and could be considered for the treatment of MG.

Since autoreactive B cells have a clear pathogenic role in the development of MG, B-cell depletion therapies, such as RTX, represent a possible solution in the management of refractory or treatment-resistant MG.2 Initially designed to treat B-cell lymphoma,9 RTX is a chimeric mouse/human monoclonal antibody directed against CD20 antigen on mature B cells which modulates B-cell activation.2 The use of RTX for MG began in 2004 when Gajra et al described a patient with both lymphoma and MG, where the treatment of the haematological condition with RTX improved MG symptoms.10 Interestingly, the beneficial effects of RTX seem different, depending on the subtype of MG: the MuSK serotype (MuSK-MG) appears to experience greater clinical benefit from treatment with RTX compared with patients with anti-AChR-Ab.1 The purpose of this article was to review the best evidence for the clinical efficacy and safety of RTX specifically in the anti-AChR-Ab+serotype of MG (AChR-MG), which is the most common (up to 90%). These data might allow a more widespread use of RTX, especially in refractory forms of generalised MG.


The authors undertook an independent PubMed literature search, employing (myasthenia)+(gravis)+(RTX) as search terms, including studies published during 20 years between September 1999 and September 2019. As the use of RTX in MG is a very common issue, the analysis was confined to studies involving at least five patients with confirmed generalised AChR-MG. Review articles, editorials, viewpoint and consensus papers were considered to screen for new records. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses diagram describes the search from literature (figure 1); 172 articles were identified through PubMed search and 7 additional records were found through other sources; 179 records were screened, from which 89 were excluded by reading the abstract, and 90 full-text articles were assessed for eligibility; finally, 13 articles (4 open-label trials and 9 retrospective studies) fulfilled the established criteria, so they were studied in depth.

Figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses diagram describing the search from literature; 179 records were screened, from which 13 articles (four open-label trials and nine retrospective studies) were selected. AChR, acetylcholine receptor; MG, myasthenia gravis.


Online supplementary file 1 summarises data from selected studies on the use of RTX in AChR-MG. Thirteen studies have been selected, showing good safety.1 2 6 7 9 11–18 The data collected from each report included the number and regimen of infusions used for therapy with RTX, antibody type and titres pretreatment and post-treatment, follow-up duration, data on safety and postintervention status of MG. The quality of data and the measures of outcome were heterogeneous among the studies included, and in many instances, some elements of data were lacking.

Supplemental material

Number of patients, RTX dose and frequency

A total of 165 patients with AChR-MG from 13 studies were treated with RTX with a follow-up ranging from 6 to 84 months. Posology and administration schemes were heterogeneous among studies ranging from a minimum of two to a maximum of three cycles of induction, followed by a maintenance regimen.

The most commonly used induction regime was 375 mg/m2 weekly for four consecutive weeks (total 1.5 g), followed by infusions of 1000 mg given twice with an interval of 2 weeks (total of 2 g). There were also studies including very low doses of RTX (375 mg/m2 twice with a 2-week interval and 100+500 mg/m2). Maintenance regimens were provided in case of relapse with doses of 375–750 mg/m2.

Side effects

RTX was well-tolerated in most patients. The most common complaints were infusion reactions with warm sensation or paraesthesia, flu-like syndrome, headache, flushing, rigours, chest discomfort and pruritus. Infectious diseases due to immunosuppression and neutropenia have been reported, including respiratory tract infections, erysipelas, gastroenteritis, reactivation of herpes zoster and giardiasis. Progressive multifocal leucoencephalopathy (PML) and cholecystitis are the most severe complications described in a minority of patients (2/165, 0.01%). In a single case, a patient developed worsening myasthenic symptoms following RTX infusion.


The measures of outcome were heterogeneous among the studies included, but RTX led to an overall sustained clinical improvement with prolonged time to relapse, in parallel to a reduction or discontinuation of other immunosuppressive therapies. A benefit from RTX treatment was measured with the myasthenic muscle score or manual muscle testing (MMT) in the remaining studies. However, beneficial effects of RTX with a significant clinical improvement were reported in 68% of patients (113/165).


A full remission of MG was reported in a study of 14/39 patients (36%).

Minimal manifestation (MM) status

Four studies expressed the benefit from therapy with the percentage of patients reaching a MM status; among these studies, the percentage of patients wtih AChR-MG reaching MM status ranged between 27% and 64% (mean 54%).

Reduction of immunosuppressant burden

In nine studies, a significant reduction of the immunosuppressant burden was achieved in most patients after treatment with RTX. In particular, seven studies demonstrated a reduction in the average dose of prednisone (many studies reporting a tapering to 50% or less), while three studies reported a decrease in the number of required IVIg and plasma exchange sessions.

Acetylcholine receptor (AChR) antibody titres

AChR antibodies titres were monitored in eight studies. In seven studies, AChR antibodies titres decreased after RTX treatment, but in one study, the antibody levels were variable and did not decrease significantly. Moreover, in all studies reporting a decrease in antibody titres, the reduction did not correlate with the clinical outcome.


The presence of thymoma in AChR-MG was specified in eight studies, accounting for 34 out of 165 overall patients with AChR-MG (online supplementary file 1). Some patients with thymoma did not respond at all to RTX, resulting unchanged or showing only a limited improvement7 14–16; others reported a significant improvement.7 9 11 12 15 16 However, the presence of thymoma has not been associated with a different response to RTX in AChR-MG.7 15 16


The data obtained over the last few years on the use of RTX in refractory MG are heterogeneous in terms of posology, administration scheme and patients’ evaluation. However, previous studies have shown that RTX is safe, since the frequency of serious adverse events is low.2 6–10 The most common side effects are related to the infusion and can be easily prevented by pretreatment with steroids and antihistamines.1 2 9 To our knowledge, only one case of PML has been reported in a myasthenic patient treated with two lines of immunosuppressive drugs: azathioprine over 4 years followed by mycophenolate mofetil for 6 years, concomitant with RTX (total dose: 10 g).7 Some patients developed aseptic leucopenia2 11 while others experienced a reactivation of latent infections such as herpes zoster or giardiasis.9

A clear improvement is well documented for AChR-MG, with ongoing remissions of more than 1 year after treatment.1 2 6 7 Despite this, the efficacy of RTX in AChR serotype is still debated.2 11–13 Fewer reports are available for AChR-MG, because it usually has a better response to common therapies. So, there could be a reporting bias toward successful treatment in AChR-MG, only because the use of RTX in this population is more frequent.9 Hence, several studies question the efficacy of RTX in AChR-MG, especially in comparison to MuSK-MG.14–16 There are only four prospective open-label studies which include a low number of patients.6 12 17 18 The others are small retrospective studies and case series. What emerges from their analysis is that RTX leads to a sustained clinical improvement with prolonged time to relapse, in parallel to a reduction or discontinuation of other immunosuppressive therapies. The dosage of RTX in the management of exacerbations, induction therapy and maintenance has not been established. In some studies, patients underwent a minimum of two cycles regardless of clinical status, while in others subjects underwent a supplementary cycle only in case of relapses.1 9 11–15

Some authors also found a decrease in anti-AChR-Ab levels.1 2 11 16 17 However, in some cases, the reduction in antibody titres does not correlate with the clinical outcome,12 14 16 17 maybe because autoantibodies may accumulate at the neuromuscular junction.6 Another possible explanation is that RTX is responsible for the removal of B cells, rather than autoantibodies. Anderson et al evidenced early depletion of CD19/CD20-positive cells in the course of treatment by serial monitoring of lymphocyte counts.18 B-cell depletion has a better correlation with the clinical response; for example, Blum et al noted B-lymphocyte recovery in all four patients who had a relapse and required further therapy.9 Hence, monitoring B-lymphocyte percentage would be more helpful than monitoring autoantibody levels in establishing when RTX retreatment is necessary, and this biomarker has been used as a powerful tool in a recent study.15

In the analysed studies, RTX seems to be more effective in anti-MuSK-Ab+MG (MuSK-MG) (online supplementary file 1). For example, Topakian et al reported an outcome of remission in in 35.9% (14/39) of AChR-Ab+ patients and MM or better in 64.1% (25/39).16 In Beecher et al, patients with AChR-MG showed a mean MMT reduction from 10.3±5.1 to 5.5±2.6 (p=0.018), whereas patients with MuSK-MG demonstrated a mean MMT decrease from 10.0±3.6 to 1.1±2.0 (p<0.0001).6 Blum et al described a clinical remission in all patients with MuSK-MG, while 3/11 patients with AChR-MG had MM status and another 5/11 had functional improvement with persistent signs and symptoms of MG.9

Different responses to RTX in anti-AChR and MuSK subtypes of MG may be explained by divergences in their pathogenesis. Indeed, IgG1 and IgG3 are the main antibody subclasses in AChR-MG, whereas in MuSK-MG, the prevalent subtype is IgG4.13 There are also different subsets of T-helper cells: T-helper 1 response is related to IgG1 and IgG3 generation, while T-helper 2 cells are responsible for the production of IgG4.13 Moreover, IgG4 is almost exclusively produced by short-lived plasma cells (PCs), thus explaining the preferential efficacy of RTX in patients with MuSK-MG.17 Indeed, RTX targets CD20 antigen expressed by pre-B-cell stage before differentiation to PCs and depletes short-lived PCs while having little or no effect on the long-lived PCs subset.17

A phase II, randomised, multicentre, placebo-controlled, 52-week clinical study which evaluates the efficacy, safety and pharmacodynamics of RTX in MG is still ongoing. It will probably clarify if RTX may represent a valid alternative treatment in patients with AChR-MG who have inadequate responses to other therapies.

Limitations and future directions

The majority of the studies analysed presented many pitfalls that could have made it difficult to evaluate the efficacy of RTX in AChR-MG: first of all, the measures of outcome and the grade of involvement among different studies are heterogeneous. Though a well-documented efficacy is available to date for some clinical instruments, such as Myasthenia gravis activities of daily living and quantitative myasthenia gravis scores, less powerful evidence is available for other indexes such as MMT, because of the fluctuations of motor signs in MG.19

A further limitation emerges from the inhomogeneity in the treatment in patients treated with RTX, as, for ethical reasons, the majority of patients underwent other treatments, including thymectomy, during the studies.9 It is a well-known fact that thymectomy can impact on the severity of AChR-MG also after a long time,20 and this issue might have influenced the interpretation of results from the mentioned studies.

Finally, the number of cycles and administration schemes of RTX is broadly variable among studies, and sometimes those can be very different in the same study.

Moreover, the two different serogroups, MuSK- and AChR-MG, which are very different entities of the same disease for pathophysiology, are generally treated with different dosages in the same studies. Patients with MuSK-MG generally have an earlier onset, with bulbar involvement compared with patients with AChR-MG. B cells producing a different antibody have different characteristics: they differ in the class of antibody (IgG1, IgG2 or IgG4 in MuSK-MG) and in the life span of memory cells, which is longer in the AChR subtype of MG; these data suggest that a different and individualised dosage, taking into account the clinical assessment and the levels of CD19/CD20 cells, could enhance the efficacy of RTX in AChR-MG.


Previous studies have shown the benefits of RTX treatment in refractory generalised MG by determining B-cell depletion, clinical improvement and reduction of immunosuppressant burden. The effects of RTX are more evident in MuSK-MG. On the contrary, data about AChR-MG are not univocal, because the majority of studies still have poor internal validity and included a small number of patients. Randomised controlled trials are needed to assess the effect of RTX in these subgroups of patients with MG. The number of cycles and administration schemes of RTX have not been defined yet. Future studies should take into account the subtype of MG with different and individualised dosages for MuSK-MG and AChR-MG. Finally, a universal and reliable measure of outcome and severity of the disease should be employed to allow an easier interpretation of the results among studies on RTX.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors VDS and AL planned the study. MGR contributed to the study conception and design. The literature search and data analysis were performed by VDS, AL and MGR. The first draft of the manuscript was written by VDS, AL and MGR. CR, ADM and FB revised the work. VDS submitted the study.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.