Eur Neurol 2005;53:146–150
Received: December 9, 2004
Accepted: March 22, 2005
Published online: May 17, 2005
Efficacy of Low-Dose FK506 in the
Treatment of Myasthenia gravis –
A Randomized Pilot Study
Yuriko Nagane Kimiaki Utsugisawa Daiji Obara Ryushi Kondoh
Department of Neurology, Iwate Medical University, Morioka , Japan
FK506 Interleukin-2 Myasthenia gravis
Peripheral blood mononuclear cells
To determine the efficacy of low-dose FK506 in the treatment
of myasthenia gravis (MG), untreated de novo patients
were randomly selected to receive treatment with
(n = 18) or without (n = 16) FK506, and were evaluated
for 1 year after treatment with limitation of daily dose of
prednisolone. Low-dose FK506 reduced the duration of
early-phase therapy in hospital (p ! 0.05) and the need
for combined therapy with plasmapheresis and highdose
intravenous methylprednisolone or high-dose intravenous
methylprednisolone alone (p ! 0.05). It also
reduced the daily dose of prednisolone (p ! 0.05) required
to maintain minimal manifestations of MGFA
postintervention status. None of the patients exhibited
significant side effects up to 1 year after treatment. These
findings suggest that low-dose FK506 is safe and efficacious
for the treatment of de novo MG patients.
Copyright © 2005 S. Karger AG, Basel
Myasthenia gravis (MG) is an autoimmune disease
generally mediated by antibodies against the acetylcholine
receptor (AChR) of skeletal muscle (AChR Ab)  .
Production of these antibodies in B cells depends upon
AChR-specific T cells  . Although both Th1 and Th2
cytokines generally play important roles in the pathogenesis
of MG  , T-cell proliferation and activation are apparently
induced by interleukin-2 (IL-2), particularly in
severe cases of MG  . Treatment with the IL-2 fusion
toxin DAB 389 IL-2, which is toxic to cells with high-affinity
IL-2 receptors, prevents experimental autoimmune
MG  . These findings suggest that the IL-2-mediated
immunoresponses in MG arise from functional abnormalities
of T cells. The immunosuppressant FK506 binds
to a cytoplasmic protein (FK binding protein) and inhibits
calcineurin  . The principal biological effects of calcineurin
inhibition include a decrease in antigen-stimulated
IL-2 production in T cells and a decrease in IL-2
receptor expression on T cells [6, 7] . FK506 has been
proven safe and effective in preventing organ rejection in
clinical organ transplantation  . Furthermore, in noncontrolled
studies, it has been observed that even a low
dose of FK506 can be used as an effective supplementary
treatment for MG and produces no significant side effects
[9–11] . In the present study, we examined the efficacy of
low-dose FK506 for treatment of de novo MG patients
in a randomized prospective clinical trial.
Fax +41 61 306 12 34
© 2005 S. Karger AG, Basel
Accessible online at:
Department of Neurology, Iwate Medical University
Morioka 020-8505 (Japan)
Tel. +81 196 51 5111, Fax +81 196 54 9860, E-Mail email@example.com
Subjects and Randomization
Thirty-six de novo MG patients between the ages of 21 and 82
years (mean age 55.4 8 16.4 years; 10 males and 26 females) were
consecutively enrolled into the study, and were randomly and reciprocally
selected to receive treatment with or without FK506.
However, the present study design was unblinded and nonplacebo
controlled. Informed consent was obtained from all subjects. FK506
(Fujisawa Pharmaceutical Co. Ltd., Osaka, Japan) was administered
orally at a daily dose of 3 mg, which is approximately 20% of
the dose given initially to organ transplant patients  . The diagnosis
of MG was based on widely accepted criteria  . Factors for
exclusion from the present study were previous immunosuppressant
therapy, elevated levels of serum liver enzymes, renal insufficiency,
pregnancy and age younger than 20 years.
Study Design and the Primary Endpoints
The patients were admitted to our hospital for early-phase therapy,
which was started with FK506 and/or prednisolone (PSL) administration.
For patients older than 50 years, thymectomy was not
strongly recommended if there were no signs of thymoma. In patients
who received thymectomy, no pharmacological intervention
was performed before thymectomy, and administration of FK506
and/or PSL was started within 2 days after thymectomy. Normal
quality of life corresponding to the category ‘Minimal Manifestations’
of Myasthenia Gravis Foundation of America (MGFA)
Postintervention Status (MM)  was achieved in both groups of
patients by administering the following treatment as required.
Treatment in the early-phase therapy during hospital stay consisted
of oral PSL not exceeding a daily dose of 20 mg, and combined
therapy with plasmapheresis (immunoadsorption method  ; IA)
and high-dose intravenous methylprednisolone (HMP) (IA + HMP)
 . HMP was performed by serial intravenous injection of 1.0 g
methylprednisolone/100 ml saline immediately after IA and on the
morning of the following 2 days  . Anticholinesterase medication
was prohibited in the early-phase therapy. PSL dose was tapered
and adjusted to the minimal dose to maintain MM.
The endpoint of the early-phase therapy was defined as the time
at which clinical symptoms remained MM with PSL not exceeding
10 mg per day. The efficacy of FK506 was evaluated by comparison
of the frequency of IA + HMP and the duration of early-phase
therapy between the two groups of patients treated with or without
FK506 (see table 2 ).
Follow-Up Phase Therapy
After discharge from the early-phase therapy in hospital, the
patients were followed up and treated to maintain MM for 1 year
after the beginning of treatment. In the follow-up phase therapy,
IA + HMP, HMP alone, oral PSL (not exceeding a daily dose of
20 mg) or pyridostigmine bromide (60 or 120 mg per day) was administered
as needed to maintain MM.
The endpoint of the follow-up phase therapy was defined as 1
year after the beginning of treatment. Efficacy of FK506 was evaluated
by comparison of the number of treatments with IA + HMP
or HMP alone during the follow-up phase and the dose of oral PSL
per day required at the endpoint between the two groups of patients
treated with or without FK506 (see table 2 ).
Table 1. Profile of patients treated with and without FK506
Type of therapy
(n = 18)
(n = 16)
Age, years 56.6817.0 54.2816.1
Gender, m/f 5/13 4/12
Time since onset, months 11.587.7 10.5813.5
AChR Ab-positive cases, % 72 (13/18) 50 (8/16)
Thymectomy, % 39 (7/18) 38 (6/16)
Thymic histology, %
Classification a , %
Severity score b 8.584.6 9.386.5
No significant difference was observed.
MGFA Clinical Classification .
MGFA Quantitative MG score .
The protocol for this study was approved by the ethics committee
of our institute.
Clinical classification was performed according to MGFA 
( table 1 ). Clinical severity was determined by patients and participating
physicians (Y.N., D.O. and R.K.) (not blinded to the treatment
type) according to MGFA quantitative MG score  ( table
1–3 ). Each patient was assessed by the same physician throughout
the course of the present study. The ratios (posttreatment to
pretreatment) of clinical severity score were calculated. Severity
score and severity ratio were estimated for individual patients before
and at 1 month, 6 months and 1 year after the beginning of
treatment ( table 3 ).
Blood sampling and examination of IL-2 production by peripheral
blood mononuclear cells (PBMs) were performed as described
elsewhere  . In brief, PBMs from the patients were cultured in
standard RPMI medium supplemented with 1% fetal bovine serum
and 20 g/ml of phytohemagglutinin for 48 h. Then, the IL-2 concentration
(pg/ml) of the supernatant was measured by solid-phase
enzyme-linked immunoassay (ELISA)  . Serum AChR Ab titer
was estimated by the immunoprecipitation method using 125 I-alpha-bungarotoxin.
The ratios (posttreatment to pretreatment) of
PBM IL-2 production and AChR Ab titer (for patients with elevated
AChR Ab titer) were calculated. PBM IL-2 production, PBM
IL-2 production ratio and AChR Ab ratio were estimated for individual
patients before and at 1 month, 6 months and 1 year after
the beginning of treatment ( table 3 ).
Effects of FK506 on MG Eur Neurol 2005;53:146–150 147
Table 2. Comparison of the items for estimation at the endpoints (early phase and follow-up phase) in patients
treated with and without FK506
Type of treatment
(n = 18)
(n = 16)
Total early-therapy period, weeks 3.781.2* 5.882.9
Severity score a /severity ratio after total early therapy 1.781.3/0.1980.13 2.282.0/0.2380.23
Number of IA + HMP treatments 1.381.5* 3.182.3
Dose of PSL at the endpoint, mg 5.184.1 6.783.0
Follow-up phase therapy (up to 1 year)
Number of IA + HMP treatments 0.280.5* 1.181.6
Number of HMP treatments 0.480.9* 1.882.7
Dose of PSL at the endpoint, mg 4.684.1* 8.182.6
* p < 0.05 (Mann-Whitney U test) compared with patients treated without FK506.
MGFA Quantitative MG score .
Table 3. Comparison of clinical parameters after beginning of therapy in patients treated with and without FK506
Type of treatment
with FK506 (n = 18) without FK506 (n = 16)
Severity score a 8.584.6 9.386.5
IL-2 production, pg/ml 1,036.28647.1 1,167.98808.1
One month after beginning of treatment
Severity score a /severity ratio 2.981.5***/0.3880.19*** 3.983.8**/0.5580.47*
IL-2 production, pg/ml/IL-2 production ratio 530.88193.7***/0.6680.30** 874.68566.3*/0.8480.48
Anti-AChR antibody ratio 0.9780.30 b (n = 13) 0.6680.29* (n = 8)
Plasma FK506, ng/ml 6.681.2
Six months after beginning of treatment
Severity score a /severity ratio 1.981.6***/0.2280.17*** 2.383.1*/0.2180.23*
IL-2 production, pg/ml/IL-2 production ratio 495.88215.2** ,b /0.6280.38** ,b 890.68635.4*/0.9080.48
Anti-AChR antibody ratio 0.6580.28** (n = 13) 0.3880.26** (n = 8)
Plasma FK506, ng/ml 6.781.4
One year after beginning of treatment
Severity score a /severity ratio 1.981.6***/0.2480.18*** 1.981.6*/0.2380.13*
IL-2 production, pg/ml/IL-2 production ratio 528.88250.3** , b /0.6480.32** ,b 925.08622.6*/0.9880.50
Anti-AChR antibody ratio 0.5980.28** (n = 13) 0.4380.29** (n = 8)
Plasma FK506, ng/ml 6.681.4
* p < 0.05; ** p < 0.01; *** p < 0.001 (Wilcoxon signed-rank test) compared with values before treatment.
MGFA Quantitative MG score .
p < 0.05 (Mann-Whitney U test) compared with patients treated without FK506.
Eur Neurol 2005;53:146–150
Differences in the evaluation items of the efficacy of FK506
between the two groups were estimated with the Mann-Whitney U
test. Changes in clinical and laboratory parameters in each patient
group at 1 month, 6 months and 1 year into the treatment (compared
with before treatment) were estimated with Wilcoxon signedrank
test. The significance level was set at p = 0.05.
A side effect questionnaire was completed by the patient at each
visit. Blood test results were monitored by the nonblinded doctor.
Two patients selected to receive treatment without
FK506 were excluded from the study because daily dose
of PSL exceeded 20 mg to achieve MM in the early-phase
therapy. The final analysis was performed for 18 patients
treated with FK506 and 16 patients treated without
FK506. There was no difference in background, including
the frequency of thymectomized patients between the
two groups ( table 1 ).
The Primary Endpoints
At the endpoint of the early-phase therapy ( table 2 )
and at 6 months and 1 year into the treatment ( table 3 ),
MM was achieved in both groups of patients, and the
mean severity score and the mean severity ratio were significantly
decreased in both groups of patients, compared
with before treatment, with no difference between the two
groups ( tables 2 and 3 ).
The number of treatments with IA + HMP was significantly
lower in the patients treated with FK506 than
in those treated without FK506 during early-phase therapy
(p ! 0.05; table 2 ). The period of early-phase therapy
was significantly shorter in the group treated with FK506
(p ! 0.05; table 2 ). During the follow-up phase therapy,
the number of treatments with IA + HMP and the number
of treatments with HMP alone were significantly lower
for the patients treated with FK506 than for those treated
without FK506 (p ! 0.05 and p ! 0.05, respectively;
table 2 ). At 1 year into the treatment, the PSL dose required
was significantly lower for the patients treated
with FK506 than for those treated without FK506 (p !
0.05; table 2 ).
At 6 months and 1 year into the treatment, PBM IL-2
production and PBM IL-2 production ratio were more
markedly decreased in the patients treated with FK506
than in those treated without FK506 (p ! 0.05; table 3 ).
Although the mean AChR Ab ratio was significantly
decreased in the patients treated without FK506 at 1
month, 6 months and 1 year into the treatment ( table 3 ),
it was not decreased in the patients treated with FK506
up to 6 months into the treatment ( table 3 ). The mean
AChR Ab ratio was significantly lower in the patients
treated without FK506 than in those treated with FK506
at 1 month (p ! 0.05; table 3 ), but there was no difference
between the two groups at 6 months or 1 year ( table 3 ).
FK506 was well tolerated over the 1-year treatment
period. The mean concentration of FK506 did not change
throughout the study in the patients treated with FK506
( table 3 ). None of the patients exhibited significant side
effects up to 1 year into the treatment. Only 1 patient who
had been suffering from hypertension exhibited increased
levels of serum creatinine (1.4–1.5 mg/dl; normal range
In the present study, we demonstrated that using lowdose
FK506 to treat de novo MG patients reduces the
number of treatments with IA + HMP or HMP alone,
reduces duration of the early-phase therapy period in hospital,
and reduces the dose of PSL at 1 year into the treatment.
None of the patients exhibited significant side effects
up to 1 year into the treatment, which is consistent
with previous reports [9–11] , probably because of the
low-dose administration (daily dose of 3 mg) of FK506
 . Furthermore, among the 18 patients treated with
FK506, 4 patients (22%) maintained MM sufficiently
with FK506 alone during the follow-up phase (severity
score: 6.2 8 3.3 before treatment; 1.3 8 1.1 at 1 year).
This indicates that low-dose FK506 is safe and efficacious
for treatment of de novo MG patients. Among the 4 patients
who maintained MM sufficiently with FK506
alone, 3 exhibited elevated PBM IL-2 production
( 1 1,250 pg/ml  ) before treatment, generalized symptoms
of MG and elevated serum AChR Ab titer. This appears
consistent with previous reports that MG patients
with elevated PBM IL-2 production exhibited immediate
effects after FK506, with clinical characteristics including
generalized symptoms of MG and serum AChR-Ab
positivity [11, 15] .
The therapeutic protocol in the present study would
be somewhat different from an orthodox method that attaches
more importance to oral steroid therapy. However,
Effects of FK506 on MG Eur Neurol 2005;53:146–150 149
given that IA + HMP can be safely applied not only to
overcome acute deterioration but also to reduce the dose
of oral PSL and aim at remission in MG patients  ,
that the number of treatments is simple for evaluation,
and that low-dose PSL is the preferred treatment to avoid
side effects, the present methods can be considered for
treatment or clinical trial of MG.
The frequency of thymectomy was relatively low in the
patients with and without FK506 (39 and 38%, respectively;
table 1 ), possibly due to their relatively higher age
(56.6 8 17.0 and 54.2 8 16.1 years, respectively; table 1 ).
Although no significant difference in the effects of lowdose
FK506 was observed between thymectomized
(n = 7) and nonthymectomized (n = 11) patients at up to
1 year (not shown), longer-term observation is needed to
investigate the effects of low-dose FK506 on the need for
In the patients treated with FK506, significant amelioration
of symptoms was observed at 1 month into the
treatment with FK506. However, serum AChR Ab titer
was unchanged at 1 month and then gradually decreased
at 6 months and 1 year ( table 3 ). In our previous study,
despite amelioration of symptoms, serum AChR Ab titer
was not affected in the first month after FK506 treatment
of MG  . The mechanisms of the early effect of FK506
regardless of AChR Ab titer levels remain to be elucidated.
The most recent report showed that the early effect of
FK506 probably results from its pharmacological actions
on the ryanodine receptor because of the early effects of
FK506 on MG patients with autoantibodies against the
C-terminus of ryanodine receptor  ; however, this issue
could not be addressed in the present study.
Although the present study is the first randomized trial
of the effects of FK506 on de novo MG patients, there
are limitations due to its unblinded design and the relatively
small number of patients. Larger-scale, multicenter,
double-blind and placebo-controlled clinical trials to confirm
the efficacy of FK506 in the treatment of MG are
currently in progress in Japan.
In conclusion, early administration of low-dose FK506
for the treatment of MG reduces the number of treatments
with IA + HMP or HMP alone, and reduces the
duration of the initial therapy, suggesting that it is an effective
measure of treatment for de novo MG patients.
This study was supported in part by a grant from the Ministry
of Education, Science and Culture, Japan.
1 Patrick J, Lindstrom J: Autoimmune response
to acetylcholine receptor. Science 1973; 180:
2 Fujii Y, Lindstrom J: Regulation of antibody
production by helper T cell clones in experimental
autoimmune myasthenia gravis. J Immunol
1988; 141: 3361–3369.
3 Zang G-X, Navikas V, Link H: Cytokines and
the pathogenesis of myasthenia gravis. Muscle
Nerve 1997; 20: 543–551.
4 Cohen-Kaminsky S, Gaud C, Morel E, Berrih-
Aknin S: High recombinant interleukin-2 sensitivity
of peripheral blood lymphocytes from
patients with myasthenia gravis: Correlations
with clinical parameters. J Autoimmun 1989;
5 McIntosh KR, Linsley PS, Bacha PA, Drachman
DB: Immunotherapy of experimental autoimmune
myasthenia gravis: selective effects
of CTLA4Ig and synergistic combination with
an IL2-diphtheria toxin fusion protein. J Neuroimmunol
1998; 87: 136–146.
6 Schreiber SL, Crabtree GR: The mechanism of
action of cyclosporin A and FK506. Immunol
Today 1992; 13: 136–142.
7 Kino T, Hatanaka H, Miyata S, Inamura N,
Nishiyama M, Yajima T, Goto T, Okuhara M,
Kohsaka M, Aoki H, Ochiai T: FK506, a novel
immunosuppressant isolated from a streptomyces.
II. Immunosuppressive effect of FK506
in vitro. J Antibiot (Tokyo) 1987; 40: 1256–
8 European FK506 Multicentre Liver Study
Group. Randomised trial comparing tacrolimus
(FK506) and cyclosporin in prevention of
liver allograft rejection. Lancet 1997; 344: 423–
9 Evoli A, Di Schino C, Marsili F, Punzi C: Successful
treatment of myasthenia gravis with tacrolimus.
Muscle Nerve 2002; 25: 111–114.
10 Konishi T, Yoshiyama Y, Takamori M, Yagi
K, Mukai E, Saida T, the Japanese FK506 MG
study group: Clinical study of FK506 in patients
with myasthenia gravis. Muscle Nerve
2003; 28: 570–574.
11 Utsugisawa K, Nagane Y, Yonezawa H, Obara
D, Kondoh R, Tohgi H: Effects of FK506 on
myasthenia gravis patients with high interleukin-2
productivity in peripheral blood mononuclear
cells. Muscle Nerve 2003; 27: 245–248.
12 Cosi V, Romani A, Lombardi M, Raiola E,
Bergamaschi R, Piccolo G, Citterio A, Berzuini
C: Prognosis of myasthenia gravis: A retrospective
study of 380 patients. J Neurol 1997;
13 Jaretzki A, Barohn RJ, Ernstoff RM, Kaminski
HJ, Keesey JC, Penn AS, Standers DM:
Myasthenia gravis: Recommendation for clinical
research standards. Task Force of the Medical
Scientific Advisory Board of the Myasthenia
Gravis Foundation of America. Neurology
2000; 55: 16–23.
14 Munakata R, Utsugisawa K, Nagane Y,
Yamagata M, Oikawa M, Obara D, Tohgi H:
The effect of combined therapy with immunoadsorption
and high-dose intravenous methylprednisolone
on myasthenia gravis. Eur Neurol
2002; 48: 115–117.
15 Utsugisawa K, Nagane Y, Yonezawa H, Obara
D, Kondoh R, Tohgi H: Interleukin-2 production
by peripheral blood mononuclear cells
from patients with myasthenia gravis. Eur
Neurol 2003; 49: 160–163.
16 Takamori M, Motomura M, Kawaguchi N,
Nemoto Y, Hattori T, Yoshikawa H, Otsuka
K: Anti-ryanodine receptor antibodies and
FK506 in myasthenia gravis. Neurology 2004;
Eur Neurol 2005;53:146–150
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.