Advances in the Treatment of Myasthenia Gravis

Nils Erik Gilhus


Future Neurology. 2012;7(6):701-708. 

In This Article

Immunosuppressive Treatment

Drug Treatment

A combination of corticosteroids (prednisone or prednisolone) and azathioprine remains the first-choice immunosuppressive drug alternative for most MG patients.[4,6,7,22,23] These drugs have a well-known and documented effect. Corticosteroids act after a few weeks, whereas the azathioprine effect is delayed for several months and up to a year. Corticosteroids in high doses usually have relevant long-term side effects. Thus, combining azathioprine with a relatively low, alternate day dose of prednisolone (10–60 mg) is recommended. The corticosteroid dose is usually reduced gradually. In some patients, the drug can be withdrawn. Azathioprine should be regarded as a long-term drug. When effective and without side effects, it is usually continued for years and is sometimes life-long. Azathioprine also has a dose-response curve, so that the dose can be increased and decreased according to effect and side effects. Weekly hematological monitoring is necessary during the first months of therapy as leucopenia is a feared side effect. A moderate reduction in blood cells is expected as part of the drug effect. Long-term tolerance of azathioprine is generally good. No increased occurrence of malignant disease has been proven,[5] and the drug does not increase the risk of other inflammatory disorders.[23] Infections are not reported with increased severity or frequency in MG patients treated with azathioprine and/or prednisolone.[23] MG patients with thymoma, with antibodies against titin and/or ryanodine receptor, or with previous MG-related respiratory dysfunction, usually need long-term immunosuppression.[4,6,7,22] Patients with focal or less severe symptoms, thymectomized for thymic hyperplasia and younger patients can more often reduce or withdraw the immunosuppressive drug treatment after being in a stable condition for some time. For patients with purely ocular MG, prednisolone in a low dose may me sufficient. Azathioprine and prednisolone are also used in young females. Both drugs are regarded as relatively safe during pregnancy. However, for general safety reasons, it is recommended that women plan pregnancies for periods when immunosuppression is not needed, or to withdraw immunosuppressive drugs temporarily before pregnancy if possible.[4,7,24] Prednisolone, and also IvIg, plasma exchange and acetylcholinesterase inhibitors are all recognized as safe in pregnancy.

For severe, generalized MG not responding to prednisolone and azathioprine, rituximab should be regarded as an early alternative. Rituximab is a monoclonal antibody with a direct effect on B cells. The binding of rituximab to CD20 molecules on B lymphocytes has widespread consequences for many aspects of immune function. The binding of rituximab to B cells is specific, but the resulting immune changes are widespread and much less specific. Rituximab is regarded as especially effective in antibody-mediated autoimmune disorders such as MG. In recent years, a lot of case reports and several uncontrolled patient series have been published, clearly demonstrating the disease-modifying effect of rituximab in MG.[20,25,26] The effect is also apparent for MuSK MG. Rituximab may even be more effective for this MG subgroup.[20] The main limitation for this treatment is risk of severe infection. As there is a lack of controlled MG studies, approved treatment protocols for MG do not exist. It has been suggested that rituximab can be tried in somewhat lower doses than has been customary for rheumatological disorders.[27] This may reduce the risk of side effects. Patients at risk may be defined from pretreatment screening for the presence of specific viruses. The effect on B cells can be monitored.[27] The effect on AChR or MuSK antibody concentrations does not seem to be of definite value in predicting sufficient dose or future effect. Long-term results are not yet known, including the need for long-term treatment, but a remission lasting for more than a year has been reported in many patients after two to four initial doses.

For milder MG not responding to prednisolone and azathioprine, mycophenolate mofetil should be regarded as an alternative.[4,7,23] The drug failed to demonstrate an impact in placebo-controlled studies as an add-on therapy to prednisolone.[28,29] Its advantages are few side effects, easy administration and a clinical response being evident after just a few weeks. Methotrexate and cyclosporine are alternative immunosuppressive drugs with a proven effect on generalized MG.[30,31] Tacrolimus should also be on the list of possible MG drugs.[32] A lot of new immunoactive drugs are marketed or in the pipeline, most of them being monoclonal antibodies, and they therefore bind specifically to one set of epitopes in the immune system. This does not, however, necessarily mean that they have immunospecific actions. These drugs are mostly being tested in severe and relatively common autoimmune disorders. MG is not among them. That represents a challenge; their potential benefit in MG treatment often remains unknown.


The thymus plays a key pathogenetic role in MG with AChR antibodies. This is illustrated by the clinical improvement seen after thymectomy in patients with a hyperplastic thymus. No well-controlled, randomized and prospective studies have been undertaken for thymectomy in MG, similar to what is true for most other surgical procedures. However, published evidence has led to clear recommendations of early thymectomy in patients with early-onset MG.[7,23,33] Thymus hyperplasia seen on a CT scan or MRI of the mediastinum strengthens the indication for thymectomy. Thymectomy usually has an excellent effect on childhood-onset MG and it can be safely undertaken in children as young as 5 years.[7] This indicates that the thymus exerts its major physiological role very early in life. Patients with low-affinity AChR antibodies have the same thymus pathology as those with ordinary AChR antibodies.[11] Thymectomy for this group should be undertaken according to the same criteria. The presence of cytoplasmic antimuscle antibodies in addition to anti-AChR antibodies (anti-titin and antiryanodine receptor) probably makes the improvement of MG weakness after thymectomy less likely. For those 10–15% of MG patients with a thymoma, the tumor should be surgically removed. Nearly all thymoma MG patients have anti-titin antibodies, half of them in addition antiryanodine receptor antibodies.

The effect of thymectomy on late-onset MG is doubtful. However, deciding on whether to undertake thymectomy or not from age of onset alone is not recommended. In a patient with generalized MG, for whom symptoms started at age 50–65 years, with an enlarged thymus on imaging, and antibodies against AChR only, thymectomy should be recommended.

For MG with ocular symptoms only, thymectomy is not recommended according to existing evidence.[7] For patients with anti-MuSK MG and anti-LRP4 MG, no pathogenetic significance of the thymus gland has been shown.[10,20] Thymectomy is not recommended for such patients.

As thymectomy has been an established treatment for MG for decades, and with clinical evidence strongly supporting the positive effect for defined MG subgroups, it is unlikely that a prospective, well-controlled study of unselected MG patients will ever occur. The multicenter initiative for such a study has succeeded in recruiting only a few of all available MG patients, resulting in a considerable and undefined selection.

Total thymectomy is necessary to obtain a therapeutic effect. This can be achieved either by trans-sternal access with a sternum split or with a video-assisted thoracoscopic procedure.[34] The patient should not have severe weakness with a threatening respiratory deficit at the time of surgery. Preoperative IvIg treatment or plasmapheresis is necessary in severe or moderately severe MG cases. In conclusion, thymectomy represents a well-established, safe and recommended treatment for major MG subgroups.