SAGE-Hindawi Access to Research
Volume 2011, Article ID 847393, 6 pages
Myasthenia Gravis:A Reviewof AvailableTreatment Approaches
Nils ErikGilhus,1,2Jone F.Owe,1,2JanaMidelfart Hoff,1,2FredrikRomi,1,2
GeirOlve Skeie,1,2andJohan A. Aarli1,2
1Department of Clinical Medicine, University of Bergen, 5020 Bergen, Norway
2Department of Neurology, Haukeland University Hospital, 5021 Bergen, Norway
Correspondence should be addressed to Nils Erik Gilhus, email@example.com
Received 14 April 2011; Accepted 11 August 2011
Academic Editor: Renato Mantegazza
Copyright © 2011 Nils Erik Gilhus et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
Patients with autoimmune myasthenia gravis (MG) should be further classified before initiating therapy, as treatment response
varies for ocular versus generalised, early onset versus late onset, and acetylcholine receptor antibody positive versus MuSK
antibody positive disease. Most patients need immunosuppression in addition to symptomatic therapy. Prednisolone and
azathioprine represent first choice drugs, whereas several second choice options are recommended and should be considered.
Thymectomy should be undertaken in MG with thymoma and in generalised, early-onset MG. For MG crises and other acute
exacerbations, intravenous immunoglobulin (IvIg) and plasma exchange are equally effective and safe treatments. Children and
is known in detail, but the immune therapy is still surprisingly unspecific, without a pin-pointed attack on the defined disease-
inducing antigen-antibody reaction being available.
Myasthenia gravis (MG) has a prevalence of 150 per million,
with nearly one million MG patients worldwide. The yearly
incidence is 10–15 per million per year . Before any
treatment was available the prognosis was severe, with an
expected 50% 10-years’ mortality. With modern treatment
facilities such as immunotherapy, thymectomy, and intensive
care facilities available, population-based studies show that
MG and non-MG individuals have the same life expectancy
, but still often with reduced physical abilities, reduced
quality of life, and risk of complications.
There are three key aspects of MG which define the
(i) MG is a well-defined autoimmune disease and thus
responds to immunoactive treatment.
(ii) MG is caused by impaired acetylcholine receptor
(AChR) stimulation in the postsynaptic skeletal mus-
cle membrane and thus responds to an increase in
(iii) MG has muscle weakness as the only symptom,
and consequently should respond to measures that
increase muscle function and counteract muscle
MG treatment is firmly established as the domain of
neurologists. Neurologists should be in charge even if the
target organ is skeletal muscle, disease mechanisms are
systemic, thymus is a target organ for diagnostic, therapeutic
and scientific approach, hypoventilation is a life-threatening
symptom, and diplopia often the most troublesome symp-
tom. Ten percent of MG patients have another autoim-
mune disorder in addition, further supporting the need
for complementary medical competence. Close cooperation
with other fields of medicine provides knowledge regarding
new immunoactive drugs, thus expanding the therapeutic
opportunities for MG.
For complicated and rare disorders such as MG, the
establishment of medical centres supervising the treatment
of the majority of MG patients and of all complicated
treatment experience will optimize present therapy and
2 Autoimmune Diseases
facilitate the introduction of new and better treatment
procedures. Centres with special competence and qualifica-
tions in MG treating the majority of patients will further
enhance research, including well-controlled and prospective
Ideally treatment recommendations should be based
on scientific evidence of high quality, preferentially more
than one blinded and controlled prospective study with a
sufficient number of well-defined MG patients. There are
disappointingly few such studies for MG. Recommenda-
tions therefore, rely on studies of lower quality and even
sometimes only on case reports, clinical experience, and
knowledge from non-MG treatment. It is important for
patients as well as doctors to know which treatment is
supported by high-quality evidence and which is more
tentative and based on clinical experience and circumstantial
The various subgroups of autoimmune MG respond dif-
ferently to treatment. Thus, before deciding any treatment,
all individual MG patients should be defined according to
subgroups. Classification aspects reflect the investigations of
each patient that are necessary to undertake :
(1) early-onset MG: age at onset <50 years. Thymic hy-
(2) late-onset MG: age at onset >50 years. Thymic atro-
(3) thymoma-associated MG;
(4) MG with anti-MuSK antibodies;
(5) ocular MG: symptoms only from periocular muscles;
(6) MG with no detectable AChR and MuSK antibodies;
The MG group with no detectable antibodies is het-
erogeneous. Some of these patients have low-affinity AChR
antibodies that are not detectable by the routine assays and
sometimes also thymic hyperplasia . Some may simi-
larly have undetectable MuSK antibodies, and some most
probably have autoantibodies against other antigen(s) in the
postsynaptic membrane. There are not yet any commercial
tests available for the low-affinity AChR antibodies .
MG patients with a thymoma have nearly always detectable
AChR antibodies in serum. Necessary investigations include
tests for AChR and MuSK autoantibodies and CT/MR of
the anterior mediastinum. Titin and ryanodine receptor
antibodies may be helpful for classification. For patients
with no AChR and MuSK antibodies, it is necessary with
thorough examinations to exclude other causes for their
muscle weakness, including nonautoimmune myasthenic
syndromes. Neurophysiological examinations with repetitive
nerve stimulation and jitter measurements are important to
establish the initial diagnosis, especially in patients without
MG should be classified according to severity . This is
of various interventions. An accurate MG severity evaluation
is crucial for controlled therapeutic studies. MG represents a
challenge for such evaluation due to variation among muscle
groups and variation during the day.
linked to growth and development in general and of the
immune system . The same is true for treatment of
MG women in childbearing age, mainly due to potential
effects of the disease and the therapies on the developing
child in utero . Epidemiology differs between ethnic
populations and also regarding the frequency of the various
MG subgroups . However, MG patients are classified
in the same way universally. Nonautoimmune myasthenic
syndromes (genetic, toxic) and non-MG autoimmune syn-
dromes (LEMS, neuromyotonia) are not included in this
3.Treatment of Acute Exacerbations
Acute exacerbations of MG need effective and urgent
life-saving treatment. Life-threatening hypoventilation is
the utmost threat. Plasma exchange and intravenous
immunoglobulin (IvIg) are both effective for acute MG .
Their beneficial and symptom-relieving effect is regarded
as well proven from several studies and from widespread
the clinical response is rapid, occurring already after 2-3 days
and often with a dramatic effect. This treatment should be
given for severe exacerbations and is mandatory for MG
crisis or threatening crisis. Plasma exchange or IvIg can
also be used for less severe exacerbations, before surgery or
together with initiation of immunosuppressive therapy with
a slower effect . Severe MG exacerbations with impaired
respiratory function need hospitalization and often intensive
Plasma exchange and IvIg have a similar clinical effect,
and a similar responder rate. The only controlled and
randomised study did not show any difference for these
two treatment options . Also nonrandomised evidence
favours an equal effect, although the clinical impression
may be a somewhat faster and more extensive effect for
plasma exchange. IvIg has less side effects and less severe
side effects. Optimal technique and high experience reduce
the complication rate, especially for plasma exchange. Both
procedure and may be superior from a total economic
perspective . Patients responding to plasma exchange
and IvIg are not necessarily the same. Thus, if one treatment
fails, the other may well be tried. It should be more
convenient to add IvIg after plasma exchange than doing the
procedure the other way around, this is to avoid washing
away all therapeutic immunoglobulin just given to treat the
For severe MG and in an acute situation, high-dose
parenteral corticosteroids can be given and also in addition
to plasma exchange or IvIg . An early exacerbation
can be seen after initiation of corticosteroids, but with
pharmacological doses, a therapeutic effect often appears
4.1. General. Patients with the diagnosis of MG should
always be considered for symptomatic as well as immunoac-
tive drug treatment. Nearly all patients need some treatment,
at least in periods where the disease shows clinical activity
with permanent or intermittent muscle weakness. Symp-
tomatic drugs have a short-lasting activity both regarding
effect and side effects. Dosage can be rapidly changed and
the treatment is flexible. Immunoactive drugs have an effect
linked to pathogenesis, and the effect usually needs some
time before it becomes manifest. Side effects are relevant
and should be considered in a long-term perspective.
Immunoactive drugs need special attention in children and
MGwomenof childbearing age.Thus,theconsiderations
for patient and doctor are different for symptomatic and
immunoactive drug treatment.
4.2. Symptomatic Drugs. Acetylcholine esterase inhibition
at the neuromuscular junction has a symptomatic effect
in myasthenia and especially in autoimmune MG [3, 10,
14–16]. Optimal dosage is adjusted according to effect
and side effects. Side effects appear from the nonneuro-
muscular cholinergic synapses in the autonomic system,
which are overstimulated. Alternative ways to increase the
amount of acetylcholine at the neuromuscular end plate
have been tried, but with less effect than inhibiting the
degradation. Acetylcholine esterase inhibitors have a stable
and predictable effect, apparently unchanged over years.
No scientific comparisons have been undertaken between
the various esterase inhibitors. The most commonly used
is pyridostigmine and also the faster acting neostigmine.
Ambenonium is used in some countries. Some MG patients
with anti-MuSK antibodies are hypersensitive to an increase
in acetylcholine concentration.
4.3. Immunoactive Drugs. Prednisone/prednisolone remains
a first-choice drug in MG [3, 14–18]. It has a well-
practice in a high number of patients. However, there are
no formal trials and no scientific comparisons with other
drugs. Side effects occur in most patients, and they are
usually of clinical significance. Prednisone/prednisolone is
regarded to be safe in pregnancy. To reduce the amount
of side effects, dosing the drug every second morning is
usually advocated. Most patients keep a sufficient clinical
effect on the MG symptoms with this regimen and with
markedly less side effects. Patients often continue to do well
on a very low every second day dose, but experience an
exacerbation if taking this low dose away. We recommend
cautiousness regarding MG patients doing well and being
stable on prednisone/prednisolone in a low dose; continued
long-term treatment may be necessary.
Azathioprine is the other well-established first-choice
immunoactive drug used for MG [3, 14–18]. This drug is
often used in combination with prednisone/prednisolone.
Formal scientific evidence for its effect in MG is lacking, but
a controlled trial showing the superiority of the combina-
tion prednisolone—azathioprine over prednisolone alone is
much cited . Azathioprine is regarded as safe and with
few side effects, also during long-term treatment. It is listed
among drugs that should not be used in pregnancy, although
formal evidence of teratogenic effects in MG patients is
lacking. During the first few months of treatment, the
numbers of leucocytes and leucocyte subgroups have to be
counted weekly. The clinical effect of azathioprine is slow
to appear. Improvement should not be expected to appear
until after 3–6 months, and full effect of the drug first
occurs after 1-2 years. This is a reason why azathioprine
is usually combined with other immunoactive treatments,
such as prednisone/prednisolone, and especially in the initial
phase. Marked improvement on azathioprine is reported in
70–90% of MG patients in open series.
Mycophenolate mofetil is regarded as an alternative drug
for mild MG [3, 14–18]. However, after promising results
in open MG patient treatment, randomized and controlled
trials failed to confirm a positive effect . The drug has
few and mild side effects and is easy to use both for patients
and doctors. Despite its limitations, mycophenolate mofetil
is still regarded as an alternative drug for mild MG, whereas
more severe MG is usually not treated by this drug because
of the negative controlled trials.
Methotrexate should be used only when first-choice
immunosuppressive drugs do not have sufficient effect [3,
14–18]. Methotrexate has a good and proven effect for other
autoimmune disorders, but is not formally tested for MG.
Still it should be tried in selected MG patients with a marked
functional deficit, partly because it is usually well tolerated.
Cyclosporine A is an inhibitor of T cells and has well-
documented immunosuppressive effects after organ trans-
plantation. A controlled prospective study with a limited
number of included patients proved the effect of this drug
for generalised MG . Due to the danger of side effects,
cyclosporine is regarded as a second-choice immunoactive
drug for moderate to severe MG not responding to azathio-
prine and prednisolone [3, 14–18].
Rituximab is a chimaeric monoclonal antibody that
targets B lymphocytes through its binding to the CD 20
molecule. MG is a prototype of an antibody-mediated
autoimmune disease, and so rituximab and B-cell depletion
are a very promising treatment alternative. In a recent review
by Benveniste and Hilton-Jones , the effect of rituximab
in 53MG patients was recorded, including patients with
both AChR and MuSK antibodies. The authors concluded
that markedly positive effects were observed with distinct
improvement of severe symptoms. Rituximab should be
reserved for patients with severe MG, where treatment with
prednisolone and at least two other standard immunosup-
pressive drugs has failed. For milder MG, the risk of pro-
gressive multifocal leucoencephalopathy and other potential
long-term side effects probably outweigh its therapeutic
potential. This drug seems to be particularly useful for anti-
Tacrolimus (FK506) is a calcineurin inhibitor just as
cyclosporine. The drug inhibits the proliferation of acti-
vated T lymphocytes, but also acts on ryanodine receptor-
mediated calcium release from sarcoplasmic reticulum
4 Autoimmune Diseases
in muscle cells. The drug has shown a beneficial effect in
MG, and it represents an alternative second-choice drug for
patients with ryanodine receptor autoantibodies .
Other drugs, such as cyclophosphamide and several new
and selective immunosuppressive drugs, have probably a
positive effect on MG, as they have on other autoimmune
disorders. However, this effect has not been documented so
far or is less well documented than for the above-mentioned
drugs. Potential side effects are significant.
Thymectomy should be undertaken in all the 10–15% of
MG patients with a thymoma. MG improvement some-
times occurs in such patients, but less consistently than in
patients with a hyperplastic thymus. The main reason for
thymectomy in thymoma patients is to remove a potentially
infiltrating tumour . In some patients with no or very
mild MG symptoms, a severe exacerbation of MG with an
increase in AChR autoantibodies has been reported after the
removal of a thymoma .
Thymectomy should always be considered as an early
therapeutic measure in early onset MG with generalised
symptoms [3, 10]. Many patients benefit considerably.
Thymic hyperplasia with an enlarged thymus and numerous
germinal follicles is associated with improvement after
thymectomy. Although no blinded and fully controlled
studies have been undertaken, scientific evidence and clinical
experience undoubtedly confirm thymectomy as a therapeu-
tic option . A transsternal approach and video-assisted
thoracoscopy appear equally effective . Postoperative
improvement occurs gradually after 2–24 months. Age alone
should not decide thymectomy or not. Some of the MG
patients that experience their first symptoms after the age of
50 years have a hyperplastic thymus and also other features
incommonwiththeearlyonset MGgroup.Suchpatients are
expected to respond to thymectomy.
Nonthymectomy: Late-onset MG patients should also be
considered for thymectomy, but thymectomy is undertaken
only in a minority of them [3, 10, 26]. Early debut age within
this late-onset group (<60 years) and thymic hyperplasia on
MR/CT imaging favour thymectomy. Higher age, symptoms
for a longer time period, atrophic thymus, and presence
of non-AChR antibodies against titin and/or ryanodine
receptor all count against thymectomy.
For MG with purely ocular symptoms, thymectomy
is not recommended. For MG patients with anti-MuSK
antibodies, the majority of evidence points to no therapeutic
effect of thymectomy. Anti-MuSK MG is probably not linked
to thymic pathology.
cle antibodies at repeated testing and even with generalised
myasthenic symptoms. As a proportion of such seronegative
should be an option also for patients in this group. For
patients with generalised MG, onset at relatively low age, and
with a hyperplastic thymus on imaging, thymectomy should
in our opinion be performed.
Thymectomy should always be undertaken in hospital
units with experience in this type of surgery. Neurologists
should evaluate the patients immediately before surgery
and continuously in the postoperative phase. The patients
should be in a stable condition at the time of surgery, and
the threshold for treatment with plasma exchange or IvIg
preoperatively should be low. Such treatment will secure a
fast recovery and counteract postoperative complications,
most importantly from the respiratory system.
During acute MG exacerbations, intensive care therapy with
rigorously. The marked reduction in MG mortality is for a
also pharmacological treatment and thymectomy represent
cornerstones in MG therapy.
Physical training, weight control, and sensible life style
modifications should be discussed with all MG patients
. Seasonal flu vaccination should be recommended.
ocular MG with troublesome diplopia.
MG should be treated early and with vigour, after classi-
fication of subtype and MG severity. Moderate or severe
myasthenic weakness represents an immediate and perma-
nent challenge. Treatment at an early stage with thymectomy
and/or immunoactive drugs improve long-term outcome.
With a lackof initial response, itis not sufficienttohavetried
only 2-3 alternative drug options. Drugs can be combined.
Longitudinal measurement of AChR antibodies can be
tiation between MG and non-MG symptoms experienced by
the patient . The clinical response and evaluation is most
important, but there tends to be a correlation between MG
severity/activity and AChR antibody concentration in the
individual patient. There are no studies systematically and
prospectively examining the usefulness of repeated antibody
examinations in established MG. Non-MG drugs given to
patients with MG should always be checked for potential
adverse neuromuscular effects.
Most MG patients are in the need of long-term therapy.
For patients in a stable remission when on immunoactive
drugs, a conservative policy regarding full drug withdrawal
is recommended. A low-dose prednisolone, azathioprine, or
other immunoactive drugs can in such patients be sufficient
to maintain the stable condition, but also necessary to avoid
new exacerbations. Younger patients in particular, not least
any need for continued drug treatment. Late-onset MG
patients and thymoma MG patients usually need life-long
In 10% of MG patients, the onset is before age 18 years
. The disease is very rare in infancy. In Asian populations,
up to 50% of cases present in adolescence . Most children
with MG have AChR antibodies. Those without antibodies
should be thoroughly checked for non-MG myasthenic
syndromes. The response to thymectomy is usually very
favourable, and thymectomy should be done early. Immuno-
suppressive drugs are used for moderate and severe cases in
the same way as for adults, but chronic administration of
suchdrugs in children usuallyleads to significant side effects.
Children more often obtain full remission after thymectomy,
so that withdrawal of immunosuppressive drugs should be
tried after a couple of years, especially if a marked reduction
of AChR antibody titre has occurred.
Pregnancy and giving birth for MG women is usually
(Caesarean section, forceps use) occurs more frequently
than in controls, related to prolonged labor . Anti-
cholinesterase drugs and prednisolone are considered to be
safe during pregnancy. Azathioprine and other immunosup-
pressive drugs should be withdrawn before a planned preg-
nancy and should be avoided in the fetal organ-developing
period. Arthrogryposis occurs with increased frequency in
children of MG mothers, caused by movement inhibition
in utero due to transplacental transfer of mother’s AChR
antibodies and thereby reduced function of the fetal AChR
[28, 29]. 10–15% of the children of MG mothers experience
a transient neonatal MG, usually mild and lasting only a
few days. This risk is influenced by AChR antibody subclass
and antigen specificity . Lactation is recommended by
most neurologists irrespective of mother’s immunosuppres-
sive MG drug treatment. Previous thymectomy does not
influence pregnancy and giving birth negatively and could
have a protective effect on neonatal MG . Father’s MG is
not known to have any influence on the child, apart from the
increased risk for MG and other autoimmune disease due to
MuSK-antibody associated MG has a phenotype that
differs from non-MuSK MG . The patients tend to
have more severe symptoms, and the therapeutic response is
more variable. Cholinesterase inhibitors should be tried, but
report no confirmed effect of thymectomy. Immunosuppres-
sive drugs should be tried for the same indications as in
non-MuSK MG. Prednisolone/prednisone and azathioprine
have lower success rate in patients with MuSK antibody MG.
As patients with this MG subtype often have severe and
cles, effective and intense therapy is necessary. Most patients
use corticosteroids. From many case reports, rituximab and
plasma exchange seem to be important alternatives often
used, and the same is probably true for IvIg [10, 11, 14, 22].
It is a paradox that MG treatment is still so unspecific.
MG is the best characterised autoimmune disease with
well-defined pathogenetic antibodies that impair function
through the destruction and inhibition of muscle cell AChR.
Still our therapeutic immunosuppression is aimed at general
mechanisms of the immune system. The external causes
for MG are totally unknown, apart from those 10–15% of
patients with a paraneoplastic condition linked to thymic
The ultimate aim of eradicating MG by removing the
cause of the disease seems still far away.
Antigen-specific treatment of MG should be promising,
and such strategies work in animal models. Induction
of specific immunological tolerance to AChR or MuSK,
shifting the immune response from harmful to nonharmful
is theoretically possible [3, 30, 31]. Strategies involving
antigen-presenting cells are considered for treatment of
autoimmune disorders, and manipulation of this process
could be antigen-specific. T-cell receptor vaccination may
be less promising, since T-cell receptor usage is not very
restricted in human MG . A sensible approach would
be to remove the pathogenic autoantibodies specifically, or
remove the plasma cells and/or B lymphocytes producing
these antibodies . So far such treatment has not been
cost-effective. The effect has not been superior to today’s
standard treatment, costs have been higher, and procedures
more complicated. Administering nonpathogenic AChR (or
MuSK) antibodies to MG patients and thereby blocking the
action of the patient’s own pathogenic antibodies would be
an alternative experimental approach.
MG is improved by the inhibition of acetylcholine
esterase. Other non-AChR molecules could theoretically be
influenced therapeutically to improve the neuromuscular
function. So far no such additional treatment has been
established as effective in the clinical situation.
New and more selective immunoactive drugs are mar-
keted worldwide. These drugs are established as first- or
second-line treatment for an increasing number of autoim-
mune and inflammatory disorders, due to their proven
superior clinical effect. For MG, these drugs have not been
evaluated in prospective and controlled trials. The present
treatment has reasonably good effect in the large majority
of MG patients, so that most patients have a high level of
daily function and with few and modest side effects of the
treatment. However, there is a need for better and more
focused treatment. Such treatment should be established
through formal multicenter trials in MG networks, not
by random and individual off-label use of the drugs. As
pathogenesis differs in MG subgroups, the immunoactive
treatment needs to be individualised. Subgroups of MG
will respond differently to various treatment alternatives. In
the future, the detailed evaluation of each MG patient will
hopefully have distinct therapeutic consequences, so that
the treatment regime is tailored according to the specific
 J. B. Andersen, A. Engeland, J. F. Owe, and N. E. Gilhus,
“Myasthenia gravis requiring pyridostigmine treatment in a
national population cohort,” European Journal of Neurology,
vol. 17, no. 12, pp. 1445–1450, 2010.
 J. F. Owe, A. K. Daltveit, and N. E. Gilhus, “Causes of
death among patients with myasthenia gravis in Norway
between 1951 and 2001,” Journal of Neurology, Neurosurgery
& Psychiatry, vol. 77, no. 2, pp. 203–207, 2006.
6 Autoimmune Diseases Download full-text
 N. E. Gilhus, “Autoimmune myasthenia gravis,” Expert Review
of Neurotherapeutics, vol. 9, no. 3, pp. 351–358, 2009.
 M. I. Leite, S. Jacob, S. Viegas et al., “IgG1 antibodies to
acetylcholine receptors in “seronegative” myasthenia gravis,”
Brain, vol. 131, no. 7, pp. 1940–1952, 2008.
 M. I. Leite, P. Waters, and A. Vincent, “Diagnostic use of
autoantibodies in myasthenia gravis,” Autoimmunity, vol. 43,
no. 5-6, pp. 371–379, 2010.
 A. Jaretzki III, R. J. Barohn, R. M. Ernstoff et al., “Myasthenia
gravis: recommendations for clinical research standards,”
Neurology, vol. 55, no. 1, pp. 16–23, 2000.
 A. Evoli, “Acquired myasthenia gravis in childhood,” Current
Opinion in Neurology, vol. 23, no. 5, pp. 536–540, 2010.
in pregnancy and birth: identifying risk factors, optimising
care,” European Journal of Neurology, vol. 14, no. 1, pp. 38–43,
 A. McGrogan, S. Sneddon, and C. S. De Vries, “The incidence
of myasthenia gravis: a systematic literature review,” Neuroepi-
demiology, vol. 34, no. 3, pp. 171–183, 2010.
 G. O. Skeie, S. Apostolski, A. Evoli et al., “Guidelines
for treatment of autoimmune neuromuscular transmission
disorders,” European Journal of Neurology, vol. 17, no. 7, pp.
 N. E. Gilhus, “Acute treatment for myasthenia gravis,” Nature
Reviews Neurology, vol. 7, no. 3, pp. 132–134, 2011.
 I. Elovaara, S. Apostolski, P. van Doorn et al., “EFNS guide-
lines for the use of intravenous immunoglobulin in treatment
of neurological diseases,” European Journal of Neurology, vol.
15, no. 9, pp. 893–908, 2008.
 P. Gajdos, S. Chevret, and K. Toyka, “Plasma exchange
for myasthenia gravis and Intravenous immunoglobulin for
myasthenia gravis,” The Cochrane Library, no. 1, Article ID
 A. Evoli, M. R. Bianchi, R. Riso et al., “Response to therapy in
myasthenia gravis with anti-MuSK antibodies,” Annals of the
New York Academy of Sciences, vol. 1132, pp. 76–83, 2008.
 A. Jani-Acsadi and R. P. Lisak, “Myasthenia gravis,” Current
Treatment Options in Neurology, vol. 12, no. 3, pp. 231–243,
 V. Kumar and H. J. Kaminski, “Treatment of myasthenia
gravis,” Current Neurology and Neuroscience Reports, vol. 11,
pp. 89–96, 2011.
 B. M. Conti-Fine, M. Milani, and H. J. Kaminski, “Myasthenia
gravis: past, present, and future,” Journal of Clinical Investiga-
tion, vol. 116, no. 11, pp. 2843–2854, 2006.
 D. B. Sanders and A. Evoli, “Immunosuppressive therapies in
myasthenia gravis,” Autoimmunity, vol. 43, no. 5-6, pp. 428–
 J. Palace, J. Newsom-Davis, and B. Lecky, “A randomized
double-blind trial of prednisolone alone or with azathioprine
 D. B. Sanders and Z. A. Siddiqi, “Lessons from two trials of
mycophenolate mofetil in myasthenia gravis,” Annals of the
New York Academy of Sciences, vol. 1132, pp. 249–253, 2008.
 R. S. A. Tindall, J. T. Phillips, J. A. Rollins, L. Wells, and K.
gravis,” Annals of the New York Academy of Sciences, vol. 681,
pp. 539–551, 1993.
 O. Benveniste and D. Hilton-Jones, “The role of rituximab
in the treatment of myasthenia gravis,” European Neurology
Review, vol. 5, pp. 95–100, 2010.
 N. Minami, N. Fujiki, S. Doi et al., “Five-year follow-up
with low-dose tacrolimus in patients with myasthenia gravis,”
Journal of the Neurological Sciences, vol. 300, no. 1-2, pp. 59–
 K. Kondo, “Optimal therapy for thymoma,” Journal of Medical
Investigation, vol. 55, no. 1-2, pp. 17–28, 2008.
 N. E. Gilhus, J. A. Aarli, R. W. C. Janzen, H. F. Otto, E. Fasske,
and R. Matre, “Skeletal muscle antibodies in patients with
a thymic tumour but without myasthenia gravis,” Journal of
Neuroimmunology, vol. 8, no. 2-3, pp. 69–78, 1985.
 G. S. Gronseth and R. J. Barohn, “Thymectomy for autoim-
mune myasthenia gravis: an evidence-based review,” Neurol-
ogy, vol. 55, no. 1, pp. 7–15, 2000.
 R. Mantegazza, F. Baggi, P. Bernasconi et al., “Video-assisted
thoracoscopic extended thymectomy and extended transster-
nal thymectomy (T-3b) in non-thymomatous myasthenia
gravis patients,” Journal of the Neurological Sciences, vol. 212,
no. 1-2, pp. 31–36, 2003.
 J. M. Hoff, A. K. Daltveit, and N. E. Gilhus, “Myasthenia
gravis: consequences for pregnancy, delivery, and the new-
born,” Neurology, vol. 61, no. 10, pp. 1362–1366, 2003.
 A. Polizzi, S. M. Huson, and A. Vincent, “Teratogen update:
maternal myasthenia gravis as a cause of congenital arthro-
gryposis,” Teratology, vol. 62, no. 5, pp. 332–341, 2000.
 E. Fostieri, K. Kostelidou, K. Poulas, and S. J. Tzartos, “Recent
advances in the understanding and therapy of myasthenia
gravis,” Future Neurology, vol. 1, pp. 799–817, 2006.
 J. Lindstrom, J. Luo, and A. Kuryatov, “Myasthenia gravis and
the tops and bottoms of AChRs: antigenic structure of the
MIR and specific immunosuppression of EAMG using AChR
cytoplasmic domains,” Annals of the New York Academy of
Sciences, vol. 1132, pp. 29–41, 2008.
 Y. Matsumoto, H. Matsuo, H. Sakuma et al., “CDR3 spec-
tratyping analysis of the TCR repertoire in myasthenia gravis,”
Journal of Immunology, vol. 176, no. 8, pp. 5100–5107, 2006.