Article Text

Download PDFPDF

Use and monitoring of low dose rituximab in myasthenia gravis
  1. Stefan Blum1,
  2. David Gillis2,3,
  3. Helen Brown3,
  4. Richard Boyle3,
  5. Robert Henderson1,
  6. David Heyworth-Smith3,
  7. Patrick Hogan2,3,
  8. Paul Kubler1,
  9. Cecilie Lander1,
  10. Nicole Limberg4,
  11. Peter Pillans3,
  12. Kerri Prain2,
  13. Christopher Staples5,
  14. Michael Walsh3,
  15. Pamela McCombe1,
  16. Richard Wong2,3
  1. 1Royal Brisbane Hospital, Brisbane, Australia
  2. 2Pathology Queensland, Brisbane, Australia
  3. 3Princess Alexandra Hospital, Brisbane, Australia
  4. 4St Andrews Hospital, Brisbane, Australia
  5. 5Mater Hospital, Brisbane, Australia
  1. Correspondence to Dr S Blum, Department of Neurology, Royal Brisbane Hospital, Butterfield Street, Herston QLD 4029, Australia; stefan_blum{at}


Background Myasthenia gravis is an autoimmune disorder of the neuromuscular junction. Rituximab (RTX), a monoclonal antibody to CD20, leads to B lymphocyte depletion and has been used in some autoimmune disorders, including small case series of myasthenia gravis patients.

Methods A retrospective analysis was performed of all patients with acetylcholine receptor (AChR) (11 subjects) or muscle specific kinase antibody (MuSK) positive myasthenia gravis (three subjects), who had been treated with RTX in Brisbane, Australia. In most patients 1 g of RTX, in two divided doses, was given. Patients were monitored by serial clinical assessments, flow cytometry of peripheral blood B lymphocytes and antibody testing.

Results RTX led to a significant improvement in symptoms in 11 of 14 patients. Doses of immunosuppressive medications were able to be reduced in 12 of 14 patients but medications could be completely ceased in only one patient. A demonstrable reduction of autoantibody levels was found in only three AChR positive patients and one MuSK positive patient, independent of clinical improvement. Peripheral blood B lymphocyte depletion was achieved in 13 out of 14 patients. B lymphocyte recovery occurred between 9 and 30 months post RTX (median 12.3 months) and was consistently associated with worsening of clinical symptoms.

Conclusion Rituximab at a dose of 1 g appears to be beneficial in the treatment of patients with severe myasthenia gravis. Serial monitoring of peripheral blood B lymphocytes appears to be useful in guiding the need for further RTX therapy.

Statistics from


Myasthenia gravis (MG) is an autoimmune disease of the neuromuscular junction, in which autoantibodies to the acetylcholine receptor (AChR) or muscle specific kinase (MuSK) cause fatigable muscle weakness.1 Neuromuscular junction dysfunction is thought to be caused by direct binding of autoantibodies; however T lymphocytes have also been implicated in the pathogenesis.2 3 Conventional treatment of MG includes acetylcholinesterase inhibitors and a range of immunosuppressive and immunomodulatory agents. While effective in most patients, some subjects fail to improve with these therapies, have contraindications limiting their use or experience treatment related side effects.

Rituximab (RTX) is a chimeric monoclonal antibody directed against CD 20 which is found on pre-B and mature B lymphocytes, but not on stem cells or plasma cells. RTX causes depletion of B lymphocytes by cell mediated or complement dependent cytotoxicity.4 5 RTX, while initially used to treat B cell malignancies, is also of benefit in a range of autoimmune diseases, including idiopathic thrombocytopenic purpura and rheumatoid arthritis.4 6 7

In MG, there have been a number of published case reports and small case series of RTX therapy.8–13 In almost all of these reports, the treatment protocol was similar to that used in B cell lymphoma—that is, four weekly doses of 375 mg/m2.

We retrospectively reviewed all 14 patients treated with RTX for MG in Brisbane. In our series, lower doses of RTX than have been previously described (typically 1 g divided into two doses) were used. This regimen has been used successfully in other autoimmune diseases6 and has advantages in terms of cost and access. We report the results of serial peripheral blood B lymphocyte and autoantibody monitoring and discuss their relationship to both the efficacy and duration of clinical improvement.



By contacting neurologists in Brisbane, we identified all patients treated with RTX for MG between May 2006 and January 2010 in one of the three major teaching hospitals (Princess Alexandra Hospital, Royal Brisbane Women's Hospital and Greenslopes Private Hospital). The diagnosis of MG was based on clinical examination and history, electrophysiological testing and the presence of serum AChR or MuSK autoantibodies. We reviewed patient charts, letters, laboratory and electrophysiological results.

Indications for RTX included inadequate response to other therapies in 10 patients and treatment limiting side effects or contraindications to immunosuppressive agents (including recurrent urinary tract infections, frequent skin malignancies, concurrent prostate carcinoma, prior Hodgkin lymphoma, steroid induced osteoporosis, drug induced renal impairment, latent tuberculosis and hepatitis C virus infection) in four patients (patient Nos 1, 4, 7 and 9 in table 1).

Table 1

Overview of patient characteristics

Clinical data were scored using the Myasthenia Gravis Association of America (MGFA) Clinical Research Standards.14 Disease severity before therapy was determined by review of the clinical records and patient interview. Patients were assessed at 3 monthly or shorter intervals for the first 12 months after RTX and 6 monthly thereafter by their treating neurologists. Their current status was determined by clinical review by a single neurologist (SB).

Other therapies

A range of other therapies were used both before and after RTX, including pyridostigmine, prednisolone, intravenous immunoglobulin (IVIg), plasma exchange (PE) and various immunosuppressive agents (azathioprine, methotrexate, cyclosporine, cyclophosphamide, tacrolimus and/or mercaptopurine) (see table 1).


RTX was given at a dose of 1 g, independent of body surface area, usually in two divided doses of 500 mg, 2 weeks apart, in all but two patients. Patient No 5 had four doses of RTX 500 mg weekly, one day after plasma exchange. Patient No 6 had one dose of 500 mg only. In most patients at least one 500 mg dose of RTX was provided by Roche Pharmaceuticals Australia (Sydney, Australia) on compassionate grounds.

Laboratory studies

Serum AChR autoantibodies and MuSK autoantibodies were measured before and at different time points after therapy. Specimens with levels >8 nmol/l were not further diluted as the AChR assay used (RSR Ltd, Cardiff, UK) is not validated for diluted specimens.

In all patients peripheral blood lymphocyte subset analysis was performed using quantitative flow cytometry. The percentage of CD19+ B lymphocyte in the total lymphocyte population was calculated. B cell depletion was defined as B cells <1% of the total lymphocyte population whereas B lymphocyte recovery was defined as the re-emergence of ≥1% CD19+ B lymphocytes on two separate occasions after RTX. In most patients, CD19+ B lymphocyte percentages were monitored before and at 1, 3 and 6 months after RTX, with monthly testing thereafter.

Statistical analysis

The non-parametric Wilcoxon test was used to calculate differences in steroid doses before and after therapy with RTX.


Retreatment was initiated if there was recovery of B lymphocyte count over 1% on two separate occasions, together with clinical signs of relapsing disease, as ascertained by the treating neurologist. In all patients, a total dose of 1 g was administered as two divided doses of 500 mg, 2 weeks apart.

Ethics approval

This study was approved by the Royal Brisbane and Women's Hospital ethics committee. Written consent to access clinical data was obtained from all patients.


Patient characteristics

Eleven AChR autoantibody positive patients (eight women, three men) and three MuSK autoantibody positive patients (two men and one woman) were treated, with a median age of 59 years (range 22–77). The mean disease duration prior to therapy was 10.7 years (range 0.5–44). All patients suffered from generalised myasthenia. Most patients were severely affected, with two requiring invasive ventilation at the time of therapy and nine patients suffering from severe bulbar dysfunction (MGFA class V in two patients, class IVa in one and class IVb in nine). Two remaining patients were less severely affected (MGFA class IIIb in one patient and class IIa in one patient).

Clinical response and medication requirements

Of the total group, 11 patients experienced clinical improvement following RTX. Of these, three (all MuSK positive) achieved clinical remission, three further patients had minimal residual manifestations and five others had functional improvement, but still had ongoing signs and symptoms of myasthenia.

No patient has developed myasthenic crises since RTX therapy. The clinical response usually occurred within the first 2–6 weeks after administration. Patient No 6 had sustained clinical response after a dose of 500 mg of RTX only.

Three patients remained clinically unchanged but in two out of the three (Nos 1 and 7), other medications were able to be reduced. These patients included patient No 9 in whom RTX was given as monotherapy due to long standing hepatitis C, and two other patients (Nos 1 and 7) with longstanding MG.

Doses of immunosuppressive and/or immunomodulatory (IVIg, plasma exchange) therapy were reduced in 12 of the 14 patients. In one subject, all medication was ceased, while in 11, ongoing lower doses of immunosuppressive and/or immunomodulatory therapy were used. For patients taking prednisone, the average dose reduction was 52% (p=0.01, SD 18.5%).

Eight patients were receiving regular infusions of intravenous immunoglobulin, of whom a dose reduction was achieved in five (patient Nos 2, 6, 8, 10 and 11) post RTX, whereas it was continued unchanged in three and increased in one patient (mean reduction 27.5%). An additional two patients (Nos 5 and 13) were receiving regular plasma exchange prior to RTX, which was subsequently stopped in both cases.

Of the 10 patients who were given RTX because of inadequate response to other therapy, there was improved disease control in nine although the other immunosuppressive agents were continued at the same or a lesser dose.

In all four patients who were given RTX because of adverse effects to medication, cessation of one or more immunosuppressive agents was achieved. In one of these patients (patient No 7), azathioprine cessation was followed by clinical deterioration, which was successfully treated with IVIg.

Side effects

Immediate side effects occurred following RTX infusion in two patients (flu-like symptoms, warm sensation and hypertension). Patient No 8 complained of a persistent alteration of sweet taste following RTX. This patient also developed moderate eosinophilia during the first treatment with RTX, which persisted until she was treated with oral metronidazole for presumed reactivation of giardiasis. Reactivation of oral herpes zoster occurred in patient No 7 and responded to oral acyclovir.


AChR levels remained >8 nmol/l after RTX in eight patients, of whom seven improved clinically. AChR levels dropped in the other three patients, of whom two improved.

MuSK autoantibody levels were increased in one patient, decreased in a second and not measured post RTX in a third. However, all three patients achieved clinical remission (two pharmacological remission and one complete remission).

B lymphocyte response

Depletion of peripheral blood B lymphocytes occurred in 13 patients, typically by 4 weeks after the second dose. In patient No 7, the CD19+ B lymphocyte percentage was not tested until 6 months following therapy, at which stage the CD19+ B lymphocyte percentage was 2%, associated with worsening of clinical symptoms. Patient No 6 achieved peripheral blood B lymphocyte depletion by 4 weeks with a single dose of 500 mg of RTX.

Follow-up and retreatment

Patients were followed-up post RTX for a mean of 14.2 months (4–47 months). Retreatment with RTX was undertaken in four patients after 9, 12, 13 and 30 months, and patient No 5 had a third treatment 13 months after his second treatment. Retreatment was administered in these patients due to worsening of clinical symptoms, by which time B lymphocyte recovery had occurred. Clinical improvement was observed in all patients within 3 months of retreatment, including patient No 5 following his third treatment.


In this study, we performed a retrospective analysis of all 14 AChR or MuSK positive MG patients treated with RTX in Brisbane, Australia, over a 3.5 year period. All patients had severe, treatment refractory myasthenia, toxicity or contraindications to standard immunosuppressive therapy. We found clinical improvement in the majority (11 out of 14) patients following RTX, with lower doses of RTX than previously reported (ie, 1 g independent of body weight). Four patients had relapsing disease after 9–30 months post RTX, which was again successfully treated with further RTX. Interestingly, we found that recovery of B lymphocyte counts in peripheral blood occurred in a temporal relationship with clinical deterioration.

In previously reported studies, the regimen and dose of RTX was similar to that used in the treatment of B cell malignancies.8–13 Although a lower dose of RTX (generally 1 g in two divided doses) was used in our patients, clinical improvement was achieved in 11 of 14 patients, with six patients achieving remission or minimal manifestations of disease. Lower doses of RTX have also been shown to be effective in idiopathic thrombocytopenic purpura.6 A lower dose is desirable because of decreased cost and also potentially reduced toxicity, including opportunistic infections.

Similar to previous reports, patients with MuSK positive myasthenia responded very well to RTX, with ongoing remissions of more than 1 year after therapy.8 9 12 15 Fewer case reports are available for AChR positive MG,8–12 possibly due to the different severity and better response to other therapies. However, this might also indicate a reporting bias towards successful treatment (which seems to happen more frequently in MuSK positive MG). In our study, we took care to include all patients from a geographically defined area (Brisbane) into our analysis and found a clinical response to RTX in three out of three MuSK positive patients, but only in eight out of 11 AChR positive patients. Interestingly, the two patients with longstanding AChR positive MG (patient Nos 1 and 7) displayed only negligible changes in symptoms post RTX.

Previous studies in both seropositive myasthenia and other autoimmune diseases did find a decrease in autoantibody levels, but rarely their elimination.8 13 16 17 Most of our patients continued to have high levels of autoantibodies but due to the limitations of our laboratory assay, high AChR antibody (>8 nmol/l) levels could not be further assessed. Seven of our AChR positive patients had levels which were persistently >8 nmol/l, and only one out of two MuSK antibody levels were found to be decreased on repeated testing. Of note is that autoantibody levels can also remain elevated with other immunosuppressive therapies, independent of clinical response. In our practice, serial measurements of AChR autoantibody levels were unhelpful for purposes of monitoring RTX effect. Due to the small number of MuSK positive patients in our study, we are unable to comment about serial measurements following RTX therapy.

In contrast, we found that the duration of CD19+ B lymphocyte depletion had a better correlation with clinical response. B cell depletion occurred in all 13 patients in whom measurements were undertaken. B lymphocyte recovery was associated with relapse of clinical symptoms in all four patients who required further therapy. In these patients, repeat treatment with RTX again induced both clinical improvement and B lymphocyte depletion. However, in two patients B cell depletion was not associated with a noticeable clinical change, suggesting that B cell depletion is necessary, but not the only factor, for achieving clinical response.

Our interpretation of these findings is that the beneficial effect of RTX is mediated by removal of B cells rather than autoantibodies, leading to elimination of B lymphocyte antigen presentation to helper and effector T lymphocytes as well as B cell cytokine production. This has been suggested as a mechanism of action in other autoimmune diseases treated with RTX5 18 and would be consistent with evidence that the pathogenesis of MG involves T cell mechanisms in addition to antibody mediated effects.2 3 We therefore suggest that monitoring B lymphocyte percentages, in combination with clinical assessment, is more helpful than monitoring autoantibody levels in determining the timing of RTX retreatment. Our current practice is to monitor B lymphocyte percentages 4 weeks after RTX is administered (to ensure B lymphocyte depletion has occurred), followed by further testing at 3 and 6 months (as recovery of peripheral blood B lymphocytes is rare within the first 6 months). Subsequently, we increase the frequency of B lymphocyte testing to monthly in anticipation of recovery of B lymphocyte counts within 6–12 months (see table 1).

Our study is limited by its retrospective character and lack of controls. A randomised controlled trial of RTX is urgently needed, given the now numerous case reports and series. In addition, the long term effects and side effects of RTX are still to be established, and there is concern about the risk of opportunistic infections after RTX including, but not limited to, progressive multifocal leukoencephalopathy.19 However, we anticipate that obtaining long term data on the efficacy and safety of RTX by means of randomised blinded trials will be difficult and will require extended follow-up periods. National and international registries of all patients with MG treated with RTX would therefore be of additional value in providing this data.

Treatment regimens for RTX in MG need to be established and, in our opinion, should be different from those regimens used to treat B cell malignancies, mirroring the use of cytotoxic agents (ie, methotrexate, cyclophosphamide) in autoimmune neurological diseases. Serial monitoring of B lymphocyte percentages appears to be helpful in determining the timing of retreatment with RTX. The findings of this study suggest that the combination of lower doses and relatively infrequent administration of RTX can help to decrease the costs associated with RTX therapy while maintaining good clinical responses and outcomes.


View Abstract


  • Funding SB was funded by the Royal Brisbane and Women's Hospital Research Scholarship.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the Royal Brisbane and Women's Hospital Research Ethics Committee.

  • 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.