You are on page 1of 8

Molecular Genetics and Metabolism 97 (2009) 284291

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism


journal homepage: www.elsevier.com/locate/ymgme

Pharmacokinetics, safety and tolerability of miglustat in the treatment


of pediatric patients with GM2 gangliosidosis
Gustavo H.B. Maegawa a,c,d,1, Paul L.M. van Giersbergen g, Sandra Yang f, Brenda Banwell b,c,d,
Christopher P. Morgan f, Jasper Dingemanse h, Cynthia J. Tifft f, Joe T.R. Clarke a,c,d,e,*
a

Division of Clinical and Metabolic Genetics, Hospital for Sick Children, Toronto, Ont., Canada
Division of Neurology, Hospital for Sick Children, Toronto, Ont., Canada
c
Research Institute, Hospital for Sick Children, Toronto, Ont., Canada
d
Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, Ont., Canada
e
Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Que., Canada
f
Division of Genetics and Metabolism, Center for Neuroscience and Behavioral Medicine, Childrens National Medical Center, Washington, DC, USA
g
Van Giersbergen Consulting, Wuenheim, France
h
Department of Clinical Pharmacology, Actelion Pharmaceuticals, Allschwil, Switzerland
b

a r t i c l e

i n f o

Article history:
Received 3 March 2009
Received in revised form 21 April 2009
Accepted 21 April 2009
Available online 3 May 2009
Keywords:
GM2 gangliosidosis
Miglustat
Pharmacokinetics
Safety
Sandhoff disease
Substrate reduction therapy
TaySachs disease

a b s t r a c t
GM2 gangliosidosis (GM2g) is an inherited neurodegenerative disorder caused by deciency of lysosomal
b-hexosaminidase A, resulting in accumulation of GM2 ganglioside, principally in the brain. Substrate
reduction therapy is currently under investigation as a treatment. The study investigated the pharmacokinetics and safety of miglustat given as single and multiple doses in infantile and juvenile GM2g patients
for 6- and 24-months, respectively. Eleven patients with infantile (n = 6) and juvenile (n = 5) GM2g
received oral miglustat at 30200 mg t.i.d. adjusted to the body surface area. Patients underwent pharmacokinetic assessments on day 1 and at month 3. The pharmacokinetics of miglustat were described
by a 2-compartmental model with a lag time, median time to maximum concentration of 2.5 h, and terminal half-life of about 10 h. The pharmacokinetics were time-independent, and did not differ between
infantile and juvenile cohorts. The accumulation index was 1.7. Among infantile GM2g patients, the
major drug-related adverse events (DRAEs) were abdominal discomfort and atulence. In the juvenile
group, however, the major DRAEs observed were diarrhea and weight loss. One juvenile patient developed peripheral neuropathy, and others showed progression of already established neuropathy, which
was judged to be part of the natural progression of the disease. Some mild laboratory abnormalities
observed were either transient or attributable to concomitant medications. Miglustat showed similar
pharmacokinetic parameters in all patients, with no specic difference between infantile and juvenile
forms. Miglustat was shown to be a safe drug, with mild to moderate diarrhea, as an age-dependent
DRAE, which was controlled by dietary modication.
2009 Elsevier Inc. All rights reserved.

Introduction
Miglustat is a non-peptidic, water soluble, N-alkylated imino
sugar, N-butyldeoxynojirimycin (NB-DNJ; OGT-918; Zavesca).
The drug is a competitive, reversible inhibitor of UDP-glucose:ceramide glucosyltransferase (glucosylceramide synthase), the initial
step in the biosynthesis of the globo-, lacto- and ganglio-series of
glycosphingolipids [1].

* Corresponding author. Address: Division of Clinical and Metabolic Genetics, The


Hospital for Sick Children, 555 University Avenue, Toronto, Ont., Canada M5G 1X8.
Fax: +1 416 813 8520.
E-mail address: jtrc@sickkids.ca (J.T.R. Clarke).
1
Present address: McKusickNathans Institute of Genetic Medicine and Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
1096-7192/$ - see front matter 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.ymgme.2009.04.013

The administration of miglustat to patients with type 1 Gaucher


disease (GD-1; OMIM #230800), an autosomal recessive
glycosphingolipid storage disease caused by deciency of lysosomal b-glucocerebrosidase, has been shown to be associated with
improvement in hepatosplenomegaly and hematological abnormalities, as well as in quality of life [2,3]. However, diarrhea
(>80%), weight loss (approximately 60%), headache (50%), u-like
symptoms (33%) and ne hand tremor (approximately 30%) were
common. Little is known about the safety, tolerability, and efcacy
of the drug in pediatric patients, or in patients with other
glycosphingolipidoses, especially those affecting the central nervous system. Being a small molecule, miglustat is likely to cross
the bloodbrain barrier and achieve signicant levels in the central
nervous system. In fact, previous studies have demonstrated that
miglustat is measurable in the cerebrospinal uid of patients

G.H.B. Maegawa et al. / Molecular Genetics and Metabolism 97 (2009) 284291

taking miglustat with GM2 gangliosidoses (GM2g)2 [4,5]. Miglustat


has also already been used in NiemannPick disease type C (NPC;
OMIM#257220), another sphingolipidosis, most commonly caused
by deciency of a integral membrane protein involved in endosomallysosomal transport of lipids (NPC1). Improvements of specic endpoints, e.g., horizontal saccadic eye movement velocity
and ambulation, were observed in the miglustat-treated patients [6].
The GM2g (OMIM#272800, #268800, +272750) are neurodegenerative lysosomal storage diseases (LSDs) associated with massive accumulation of GM2 ganglioside in the brain caused by a
deciency in lysosomal b-hexosaminidase A (Hex A) resulting from
mutations in one of three related genes: HEXA, HEXB, or GM2A. The
disorder caused by HEXA mutations is conventionally called Tay
Sachs disease (TSD). In its most common form, it is characterized
by early onset, rapidly progressive neurodegeneration with death
in the rst 35 years of life. HEXB mutations produce Sandhoff disease (SD) and GM2A mutations produce GM2 activator deciency;
both clinically indistinguishable from TSD. Less severe mutations
in each of the genes result in later onset, more slowly progressive
disorders with variable neurological phenotypes [7]. Patients with
juvenile or adult forms of GM2g have low, but signicant levels of
residual Hex A activity [7,8].
Platt et al. have shown that miglustat decreases the rate of neuronal ganglioside storage in TSD mice [9]. When miglustat was
administered early to SD mice, the onset of neurological symptoms
was signicantly delayed, ganglioside storage was reduced and life
span was increased by 40% [10]. These results suggested that substrate reduction therapy (SRT) may be more benecial in patients
with late-onset forms of GM2g, in which some residual enzyme
activity is present or an alternative ganglioside catabolic pathway
is up-regulated (e.g., the sialidase degradation pathway for GM2),
allowing degradation of a small amount of substrate. One report
of two patients with infantile GM2g showed that treatment with
miglustat for 912 months did not arrest neurological deterioration [4]. Another case report described a child with infantile
GM2g, who received miglustat after failure of an allogeneic bone
marrow transplantation, and showed no signicant treatment effect [11]. A randomized, open-labeled study including thirty adult
patients with late-onset TSD failed to show improvements in specic outcomes, but demonstrated a safety prole consistent with
previous clinical studies in GD-1 [2,3,12].
The pharmacokinetics of miglustat in adults with GD-1 have
been reported in previous studies [2,3]; however, pharmacokinetics in children or adolescents have not been documented. We report here the pharmacokinetics of miglustat after single- and
multiple-dose administration in 6 patients with infantile and 5 patients with infantile forms of GM2g. We also report on the safety
and tolerability of miglustat treatment, based on clinical assessments, laboratory studies, and electrophysiological testing. The
efcacy results will be published separately.
Material and methods
Subjects
A total of 11 patients, 6 with infantile and 5 with juvenile GM2g,
were studied (Table 1). In each case, the diagnosis was established
by the demonstration of deciency of Hex A in peripheral blood
leukocytes or cultured skin broblasts and conrmed by identication of specic mutations in the HEXA or HEXB genes. All exhibited
signicant clinical signs of progressive neurodegeneration typical
2

Abbreviations used: DRAE, drug-related adverse event; GM2g, GM2 gangliosidosis;


GD, Gaucher disease; GI, gastrointestinal; LSD, lysosomal storage disease; NCVs, nerve
conduction velocities, NPC, NiemannPick C disease; SD, Sandhoff disease; TSD, Tay
Sachs disease.

285

of the disease, with onset between 1 and 15 years of age. At baseline, the mean age was 17.2 4.7 months for the infantile, and
14.5 5.0 years for the juvenile cohort, respectively. One infantile
patient who had previously received an umbilical cord blood transplant was 5 years at the start of the study. All had normal renal and
liver function. The patients were recruited through the physician
principally responsible for the care of the patient with parental informed consent and patients assent (if appropriate). The protocols
were approved by the institutional review boards at each institution (Hospital for Sick Children, Toronto, ON, Canada for juvenile
patients; and Childrens National Medical Center, Washington,
DC, USA for infantile patients).
Study design and treatment regimens
The study was designed as an open-label, single-arm study to
assess the pharmacokinetics, safety and tolerability of miglustat.
In order to achieve adequate levels of miglustat in the central
nervous system, the highest doses already examined in LSDs were
chosen [13]. The drug was administered at a maximum dose of
600 mg/day, divided in three doses. As the study was performed
in a pediatric population, the dose was adjusted to the body surface area (BSA) of each patient as calculated by the method described in The British National Formulary for Children [14].
Patients with a BSA less than 0.8 m2 were assigned to a dosing
regimen within the range of 5090 mg three times daily. Patients
with a BSA within 0.81.3 m2 received 100 mg three times daily,
and those with a BSA greater than 1.3 m2 received 200 mg three
times daily (Table 1). Dose regimens were modied during the
course of the study based on the frequency and intensity of
drug-related adverse events (DRAEs). In addition, to further minimize side effects, the dose of miglustat was incrementally increased from one third of the nal dose for week 1, via half of
the nal dose for week 2, to the full dose beginning from week
3 onwards. Juvenile patients were able to swallow normal
100 mg miglustat capsules. Infantile patients had drug compounded to a powder, weighed by the investigational pharmacy,
and placed in packets. Parents dissolved the appropriate amount
of drug from the packets in an aqueous solution just prior to
administration by mouth or by gastrostomy.
Pharmacokinetic assessments
Pharmacokinetic assessments were performed on day 1, at rst
exposure to the drug, and again at the third month visit (month 3)
at steady-state. Blood samples (4 mL) for the measurement of plasma miglustat concentrations were obtained immediately before
study drug administration and at 0.5, 1, 2.5, 4, 6, 9, 12, 15, and
24 h thereafter. The single morning dose of miglustat was the only
dose of the drug administered on both days of pharmacokinetic
assessments. Blood samples were collected in lithium heparin-containing tubes, mixed gently, and immediately placed on ice. Within
30 min of collection, the plasma was separated by centrifugation at
4 C, 1500g for 10 min, and transferred into polypropylene tubes.
All plasma samples were stored in an upright position at 20 C
or below. The concentrations of miglustat in plasma were determined using a validated LCMS/MS method with a lower limit of
quantication in plasma of 5.0 ng/mL; details of which have been
described previously [15].
Clinical assessments
The juvenile cohort was studied for 24 months, with clinic visits
for screening and at baseline (day 1), then at 3-month intervals for
a total of 9 visits after initiation of miglustat treatment. After one
week, and at months 2, 4, 5, 7, 8, 10 and 11, telephone interviews

286

G.H.B. Maegawa et al. / Molecular Genetics and Metabolism 97 (2009) 284291

Table 1
Demographic and dosing information of studied patients.
Patient

Age at start
(years)

Sex

Disease variant/
form

Miglustat dose
(daily) (mg)

Body weight
(kg)

BSA (m2)

Concomitant medications

HSC-102
HSC-103

8.7
16.3

F
F

SD, juvenile
TSD, juvenile

300
600a

19.3
61

0.8
1.8

HSC-104
HSC-105
HSC-107
CNMC-001

20.1
18
10.1
2.0

M
M
F
F

SD, juvenile
SD, juvenile
TSD, juvenile
TSD, infantile

600b
600b
300
180

44
47
34
8.6

1.5
1.4
1.2
0.4

CNMC-002
CNMC-003
CNMC-004

1.7
1.0
5.0

M
F
M

TSD, infantile
TSD, infantile
TSD, infantile

180
180d
270d

9.8
8.7
17.8

0.5
0.4
0.7

CNMC-005
CNMC-006

1.2
1.4

F
M

SD, infantile
TSD, infantile

180
150d

10.2
10

0.5
0.4

Propranololc, risperidonec
Alprazolamc, escitalopramc, clozapinec, minocyclinec,
aripiprazolec, haloperidol, lithium carbonate,
carbamazepine, trazodone, nezotropine, lorazepam,
clonazepam
Carbamazepinec, gabapentinc, codeine, azithromycin
Lithium carbonatec, gabapentinc, codeine
Carbamazepinec, valproic acid, clobazam
Polyethylene glycol 3350, phenobarbital,
levetiracetam, diazepam
Polyethylene glycol 3350, gabapentin, glycopyrrolate
Lamotrigine, celecoxib, levetiracetam
albuterol, ipratropium, celecoxib, docusate,
clonazepam

Levetiracetam

M, male; F, female.
a
Daily dose decreased by 300 mg after 17 months.
b
Daily dose decreased by 300 mg daily after 11 months because of DRAEs.
c
Medications in use during PK assessments.
d
Daily dose decreased by 50% after 5 weeks because of DRAEs.

were carried out with the parents of the patients to assess compliance with treatment, concomitant medications, and symptoms
possibly resulting from drug treatment. Nerve conduction studies
were performed at baseline, 6, 12 and 24 months, and 12-lead electrocardiograms (ECG) were performed at baseline, 12 and
24 months. The infantile cohort was studied for a 6-month period,
with clinic visits for screening and at baseline, then at 12 and
26 weeks. Because of the inconvenience associated with the need
to travel, the primary care provider saw patients for physical
examination, laboratory testing, and ECG at week 5. The parents
of each infantile patient were given a daily diary to record dosing,
concomitant medications, and any potential adverse events. Telephone interviews were conducted with the parents regarding these
variables at weeks 4, 8, 16, and 20.
In both cohorts, clinic visits consisted of a standardized clinical
protocol, including clinical history, general physical and neurological examinations, and review of potential DRAEs, co-morbidities
and concomitant medications. Vital signs measurements included
recording of blood pressure and pulse, temperature, respiratory
and heart rates. Standing heights and weights, or recumbent
lengths were measured in light clothing, without shoes, at every
visit.
Clinical events considered to be DRAEs were classied according to severity as mild, moderate, severe based on Common Terminology Criteria for DRAE v. 3.0 from the National Cancer Institute
[16]. DRAEs were also classied as denitive, probable, possible,
unlikely, or unrelated to study drug. Life-threatening or causing
death events that were potentially drug-related were considered
as severe DRAE.
Laboratory tests
Safety biochemical studies included measurements of plasma
electrolytes, uric acid, total protein, albumin, glucose, renal and liver function tests. Hematological and coagulation studies were
also included, as well as plasma vitamin B12 and folate levels. Urinalyses were performed using commercially available dipsticks. All
studies in juvenile patients were performed in the Department of
Paediatric Laboratory Medicine at the Hospital for Sick Children
(HSC), Toronto, ON, Canada. In the infantile cohort, the biochemical
and hematological testing was performed at the Childrens
National Medical Center, Washington (CNMC), DC, USA.

Pharmacokinetic variables
The pharmacokinetic variables of miglustat were dened as the
area under the plasma concentration time curve (AUC), the maximum plasma concentration (Cmax), the time to reach Cmax (tmax),
the elimination rate constant, kz, with the respective apparent
half-life (t1/2), and the accumulation index. The measured individual
plasma concentrations of miglustat were used directly to obtain
Cmax and tmax. AUC0t was calculated according to the trapezoidal
rule using the measured concentrationtime values above the limit
of quantication. AUC01 was calculated by combining AUC0t and
AUCextra. The AUCextra represents an extrapolated area derived from
Ct/kz, where Ct is the last plasma concentration measured above the
limit of quantication and kz was determined by log-linear regression analysis of the measured plasma concentrations of the terminal
phase. Because the apparent terminal half-life of miglustat, calculated as t1/2 = ln 2/kz, was determined over a 24-h period, i.e., only
about 2 times the t1/2, the value for this variable may have been
underestimated. The accumulation index was calculated as follows:
AUCs multiple dose/AUCs single dose, where AUCs is the area under
the curve during a dose interval s (8 h) calculated according to the
linear trapezoidal rule. The pharmacokinetic parameters were
determined on the basis of the actual blood sampling time points
using non-compartmental analysis. The pharmacokinetic evaluation was performed using the WinNonlin software package (version
3.1, Pharsight Corporation, Mountain View, CA, USA). This software
package was also used to dene a model describing the mean
plasma concentrationtime prole of all patients on day 1 and to
simulate with the obtained pharmacokinetic parameters the
multiple-dose prole.
Pharmacokinetic variables were analyzed by descriptive statistics only, and no inferential statistics were performed. The variables
were summarized as geometric means, and corresponding 95% condence interval, or for tmax as the median, including correspondent
minimum and maximum values. Covariate analysis (clinical site,
type of disease, sex) was done only graphically, displaying arithmetic mean plasma concentrations with their standard deviation.
Results
The demographics, GM2g variants and clinical forms, dose and
concomitant medications of the patients are shown in Table 1. In

G.H.B. Maegawa et al. / Molecular Genetics and Metabolism 97 (2009) 284291

some infantile patients (CNMC-003, CNMC-004, and CNMC-006),


the dose of miglustat was reduced between the two pharmacokinetic assessments. However, prior to or on the day of the second
pharmacokinetic assessment, the dose was increased and these patients received the same single dose of miglustat on both pharmacokinetic days. All patients were assessed for pharmacokinetics;
however, patient CNMC-002 completed only the rst pharmacokinetic assessment and was, therefore, excluded from all summary
statistics. The mean plasma concentrationtime proles of miglustat are shown in Fig. 1. Miglustat concentrations were slightly
lower in infantile than in juvenile cohort patients (Fig. 1A and B).
Following multiple-dose, but not after single-dose administration,
miglustat plasma concentrations tended to be higher in male versus female patients (Fig. 1C and D), and in SD versus TSD patients

287

(Figs. 1E and F). Individual pharmacokinetic parameters of all patients are summarized in Table 2. The variability of individual Cmax
concentrations was larger after single-dose administration when
compared to multiple-dose administration (Table 2, Fig. 1),
whereas the opposite was true for the variable AUC01 (Table 2).
Comparison of mean pharmacokinetic parameters obtained after
single- and multiple-dose administration revealed that the median
tmax was 2.5 h, the t1/2 about 10 h, the ratio AUC01/AUCs close to 1,
and that miglustat did not accumulate to a major extent (Table 3).
In Fig. 2, the measured and predicted miglustat concentrations
after single- and multiple-dose administration are depicted. The
selected 2-compartmental model with a lag time based on the single-dose data on day 1, predicted the multiple-dose data reasonably well.

Fig. 1. Mean plasma concentrationtime proles of miglustat after single- and multiple-dose administration in infantile (A, n = 5) and juvenile (B, n = 5) cohorts. The singleand multiple-dose analyses of the mean plasma concentrations are also depicted by sex (C and D; n = 5 for both sexes) and disease variants, TSD (n = 6) or SD (n = 4) disease (E
and F, respectively).

288

G.H.B. Maegawa et al. / Molecular Genetics and Metabolism 97 (2009) 284291

Table 2
Individual pharmacokinetic parameters of miglustat after single- and multiple-dose administration.
PK study

Patient

Cmax (ng/mL)

tmax (h)

AUC01 (ng h/mL)

1 (day 1)

HSC-102
HSC-103
HSC-104
HSC-105
HSC-107
CNMC-001
CNMC-002
CNMC-003
CNMC-004
CNMC-005
CNMC-006

2,226
1,598
2,266
1,388
2,838
2,135
2,969
1,783
2,499
1,404
949

2.5
2.5
4.0
6.0
2.5
2.5
4.0
4.0
2.5
4.0
2.5

17,909
13,127
22,931
16,534
19,272
12,399
33,699
15,584
35,358
13,547
6,577

2 (month 3)

HSC-102
HSC-103
HSC-104
HSC-105
HSC-107
CNMC-001
CNMC-002
CNMC-003
CNMC-004
CNMC-005
CNMC-006

2,328
3,438
5,656
4,276
1,471
2,032
NA
5,512
3,592
2,961
786

2.5
2.5
2.5
4.0
2.5
1.0
NA
2.5
4.0
2.5
2.5

AUCs (ng h/mL)

t1/2 (h)

Accumulation index

9.4
7.2
5.7
7.4
8.2
7.2
4.2
9.9
13.8
8.6
76
14,217
21,081
28,438
27,398
9,582
9,247
NA
19,653
21,865
15,924
4,548

6.1
14.3
8.9
5.4
9.7
18.2
NA
5.4
11.9
14.1
15.5

1.3
2.8
2.1
3.5
0.7
1.0
NA
2.4
1.7
2.3
1.2

PK (pharmacokinetics) study 1, single-dose assessment at start of treatment (day 1). PK study 2, multiple-dose assessment at month 3. Accumulation index, AUCs month 3/
AUC08h day 1; NA, not available (patient did not undergo the PK study 2).

Table 3
Descriptive statistics of pharmacokinetic parameters of miglustat after single- and
multiple-dose administration (n = 10).
Parameter

Single dose (day 1)

Multiple dose (month 3)

Cmax (ng/mL)
tmax (h)
AUC01 (ng h/mL)
AUCs (ng h/mL)
AUCs/AUC01
Accumulation index
t1/2 (h)

1822 (14362311)
2.5 (2.56.0)
15,920 (11,63521,783)

2771 (17854302)
2.5 (1.04.0)
15,167 (10,05822,869)

0.95 (0.711.3)
1.7 (1.22.5)
8.3 (7.09.8)

10.0 (7.313.9)

Data are presented as geometric means (95% condence intervals), with exception
of tmax, which is presented as Median (minmax).

Fig. 2. Simulation of the multiple-dose prole of miglustat using a two-compartment pharmacokinetic model with a lag time. Data extracted from both infantile
(n = 5) and juvenile (n = 5) patient groups. The measured data (standard deviation)
are shown by the closed triangles.

In the infantile patient cohort, abdominal discomfort along


with atulence and gassiness were the most common DRAE (Table
4). Interestingly, diarrhea, earlier reported to be one of the most

common DRAEs in the adult population, was present in only


two patients, and was mild and intermittent (Table 4). Diarrhea
was dened as an increase in frequency of bowel movements
from regular patient pattern; and loose stools was dened as
decreased stool consistency with normal frequency of bowel
movements. Signs of peripheral neuropathy, assessed clinically
by increased pain and tactile sensitivity of extremities, were suspected in one child. Episodes of irritability and some eye twitching were also observed in one child, and resulted in reduction of
the drug dose with improvement of the patients symptoms. Signs
of disease progression, such as loss of psychomotor milestones
and increased seizure frequency and severity, were noted in most
patients.
In the juvenile cohort, diarrhea was the most common DRAE,
and the mean time for the onset of diarrhea was 22 days (8
48 days) from the start of drug treatment. Loose stools were another gastrointestinal (GI) complaint frequently reported by this
patient cohort. Patients also reported vomiting, and showed
weight loss, on average 40 days (1886 days) and 2.8 months (1
6 months) after initiating the drug, respectively. In Table 5, the suspected DRAEs in the juvenile cohort are categorized according to
the likelihood of their relationship to the study drug. Among the
four patients who showed a mean weight loss of 4.8 kg (1
10.5 kg), only one recovered their original weight (HSC-107). One
patient (HSC-102) experienced a weight gain of 1.7 kg during the
study.
In general, any clinical laboratory abnormalities observed
were modest, transient and returned to normal on subsequent
assessments. Exceptions were mild elevations of sodium (n = 1),
reduced white blood cell and neutrophil counts (n = 1), and elevations of lactate dehydrogenase (LDH) (n = 1), noted in 34
safety assessments in two patients in the juvenile cohort. Four
of the juvenile patients presented mild elevations of AST, median
of 8 U above baseline that remained elevated throughout the 24month period of observation. Similar mild and consistent AST
elevations were observed in the infantile cohort. In the juvenile
cohort of patients, nerve conduction velocities (NCV) and amplitudes were reduced in the lower limbs of 3 patients before exposure to miglustat (HSC-103, HSC-104 and HSC-105). Those
patients showed progression of the neuropathy. One patient

289

G.H.B. Maegawa et al. / Molecular Genetics and Metabolism 97 (2009) 284291


Table 4
Clinical episodes suspected to be DRAEs related to miglustat in infantile GM2g patients (n = 6) categorized according to potential relationship to the drug.
Suspected DRAE

Number of
patients

Number of episodes
recorded

Pattern (episodic, intermittent,


continuous)

Severity

Denitive
Diarrhea/loose stools

Episodic

Mild

Probable
Abdominal discomfort
Flatulence and gases

2
5

14
17

Episodic
Episodic

Mild to moderate
Mild to moderate

Possible
Sudden increased sensitivity and pain (hands/feet)

Episodic

Irritability and reduced alertness


Episodes of irritability and eye twitching

2
1

8
1

Episodic
Episodic

Moderate (required drug


dose reduction)
Moderate
Severe (needed drug dose
reduction)

1
3

1
4

Continuous
Episodic

Severe
Moderate

11

Episodic

Mild to moderate

1
1
5
2

1
1
19
3

Mild
Mild
Moderate to severe
NA

Episodic
Episodic
Episodic
G-tube placement (1); dentist surgery;
G-tube re-insertion
Episodic

Unrelated
Dysphagia
Pneumonia or respiratory distress needing medical
assessment or admission
Congestion/otitis/sinusitis/
Other upper respiratory
Isolated fever
Skin infection
Seizures
Surgical or medical procedures
Viral gastroenteritis

Mild

G, gastrostomy; NA, not applicable.

Table 5
Clinical episodes suspected to be DRAEs to miglustat in juvenile GM2g patients (n = 5) categorized according to potential relationship to drug.
Suspected DRAE

Number of patients

Number of episodes
recorded

Pattern (episodic, intermittent,


continuous)

Severity

Denitive
Diarrhea
Incontinent loose stools

5
5

227
28

Episodic
Episodic

Mild
Mild

Probable
Nausea/vomiting
Stomach pain
Weight Loss

4
2
4

11
28
4

Episodic
Episodic
Continuous

Mild
Mild
Mild to moderate

Unlikely
Limb cramps
Fatigue
Sudden lower limb weakness
Epistaxis
Neuropathy
Tremors

3
3
1
1
4
2

34/week
2
4/week
3
4
2

Intermittent
Episodic
Intermittent
Episodic
Continuous
Continuous

Mild
Mild
Mild
Mild
Moderate
Mild

Unrelated
Flu-like episodes
Pharyngitis
Skin disorders
Trauma/falls
Psychiatric symptoms
Recent memory disturbances
Seizures

5
1
4
3
1
1
1

10
4
4
243
1
55
215/day

Episodic
Episodic
Episodic
Episodic
Episodic (psychosis)
Episodic
Intermittent

Drowsiness
Dysphagia
Constipation
Acute respiratory distress

1
3
1
1

1
3
33
1

Episodic
Continuous
Episodic
Episodic

Dehydration
Menstrual cycles
disturbances

1
1

1
3

Episodic
Episodic

Mild
Mild
Mild
Mild
Moderate
Mild
Moderate to severe (2
hospitalizations)
Mild
Mild to severe (one G-tube insertion)
Mild
Severe (viral upper respiratory
infection)
Moderate
Mild

G, gastrostomy.

developed NCV changes during the course of the trial; however,


the pattern was suggestive of disease progression rather than a
DRAE. The ECGs of patients in both cohorts were consistently
normal.

Discussion
The medical literature contains only two studies in which the
pharmacokinetic parameters of miglustat were reported in

290

G.H.B. Maegawa et al. / Molecular Genetics and Metabolism 97 (2009) 284291

patients with LSDs [2,3]. However, to the best of our knowledge, no


study has been conducted to assess the pharmacokinetics of miglustat in pediatric patients. We report here the rst of such a study in
a pediatric cohort of 11 patients, 6 with infantile and 5 with juvenile forms of GM2g.
Our results show that the pharmacokinetics of miglustat are
time-independent in GM2g patients as evidenced by an AUCs/
AUC01 ratio close to 1.0. Similar median tmax and apparent terminal half-life values after single- and multiple-dose administration
further support this notion. In general, the pharmacokinetics of
miglustat in GM2g patients were similar to those in adult GD patients [2,17,18]. In the present study, the terminal half-life was
slightly longer than previously reported in GD patients (56 h)
[2,3], but similar to that in healthy subjects [15]. In GD patients,
blood sampling was only done until 12 h after dosing [2], and the
present results indicate that this is not long enough to accurately
determine the terminal half-life.
The variability in exposure to miglustat observed in GM2g patients is greater than in healthy adult subjects who were administered a single 100-mg dose of miglustat [15]. This is not surprising
given the extensive age range of the cohorts studied. With increasing
age, a tendency for miglustat exposure to increase was noted (Fig. 3).
It is well known that young children may eliminate drugs faster than
adults [19], and the present results suggest that this is also the case
for miglustat. In agreement with the above nding, plasma concentrations in infantile patients tended to be lower than in juvenile
patients (Fig. 1). Covariate analysis further revealed that there were
no apparent effects of the GM2g variant (TSD or SD), but, at steadystate, males appeared to have higher miglustat plasma concentrations than females, and patients with SD higher concentrations than
TSD patients. The observed differences should be interpreted with
caution given that: (1) there were no such differences after singledose administration; (2) the small cohorts of patients included in
the study; (3) the heterogeneity of included patients with respect
to age; and (4) the observed larger variability in miglustat Cmax after
multiple dosing in comparison to single-dose administration. For the
latter nding no good explanation can be provided.
A pharmacokinetic model was established allowing relatively
accurate prediction of steady-state data based on single-dose measurements (Fig. 2). This model is the rst step in establishing a
pharmacokinetic/pharmacodynamic model linking miglustat plasma concentrations with its efcacy or safety. Such a model would
be particularly relevant for physicians regarding dose selection
when treating patients with the drug. However, clinical studies
in LSDs typically include only small numbers of patients and pharmacokinetic assessments are rarely performed.

Fig. 3. Correlation of exposure to miglustat and patients age at the start of


treatment. Individual AUCs values are shown. The solid line represents the linear
regression line, and dashed lines depict its 95% condence interval.

Potential DRAEs using miglustat have been described [2,3,20].


In the study, two serious adverse events were recorded: an emergency visit because of acute shortness of breath caused by food
aspiration and a hospitalization for seizure control in a patient
with disease-related seizures. Both events were judged not to be
drug-related (Table 5). Diarrhea and loose stools were classied
as denitive DRAEs (Table 5), and are one of the most common
DRAEs already described in the pediatric and adult patient population [2,3,21,22]. Interestingly, the frequency and intensity of diarrhea directly correlated with age. In the infantile cohort, diarrhea
was not the most common DRAE; however, atulence, and abdominal discomfort were frequently observed (Table 4). The frequency
and intensity of these adverse events decreased by dietary manipulations reducing the intake of carbohydrates, as well as measures
to ensure the drug was taken during fasting. In juvenile patients,
the frequency of diarrhea, atulence, and cramps was decreased
by adhering strictly to a lactose-free, low carbohydrate diet, along
with oral administration of lactase before meals. Some patients
were also treated with loperamide. The two oldest patients (HSC104 and HSC-105) experienced the most severe diarrhea, and after
month 11, the dose of miglustat was reduced by 50%, which resulted in partial resolution of GI complaints. These patients were
the least physically and neurologically impaired, and were self-sufcient for feeding and meal-decisions, making dietary control
difcult.
Weight loss, not observed in the infantile cohort, occurred in 4
juvenile patients (Table 5), and was classied as a probable DRAE.
Previous studies had reported weight loss as a frequent DRAE in
patients on miglustat; however, it was transient and reversible
[2,3,22,23]. We noted that older patients had the greatest weight
loss, HSC-103 (16.3 years) and HSC-104 (20.1 years) lost 10.5 and
7 kg, respectively. Patient HSC-107 (10.1 years) showed a minimal
weight loss of 1 kg. No correlation was observed between severity
of diarrhea and weight loss. Weight loss did correlate with the progression of neurological disease seen in these patients. Peripheral
neuropathy, previously reported as a suspected DRAE in 4 patients
[2,3], was classied as an unlikely DRAE according to the scale used
in this study (Table 5). Patient HSC-107 showed decreased NCVs
and amplitude in the lower limbs. Peripheral neuropathy is a common feature of GM2g as the disease progresses, making the assessment of a possible drug relationship difcult. One infantile patient
developed hypersensitivity of his hands and feet, possible signs of
neuropathy, which improved with a 50% decrease in the dose of
miglustat (Table 4). Two juvenile patients from the study reported
tremors, thought to be disease-related. One of them had tremor before initiation of the study and improved after a concomitant medication was discontinued.
In general, laboratory abnormalities were mild and transient.
Nevertheless, 4 juvenile patients (HSC-103, HSC-104, HSC-105 and
HSC-107) showed persistent, mild elevations of AST. In three of them
(HSC-103, HSC-105 and HSC-107), this elevation was present at
baseline. All patients in the infantile cohort also showed mild AST
elevations (44.0 4.3 U/L; normal range <36 U/L). On the other
hand, ALT measurements in these patients were consistently normal. AST may reect disturbances in hepatocytes, as described in another LSD disease model [24], or, interestingly, may reect neuronal
impairment typical in GM2g. Satoh et al. recently reported signicant elevation of AST, along with LDH, in cerebrospinal uid (CSF)
of a dog model of GM1 gangliosidosis, suggesting their use as biomarkers of this condition [25]. One of us (CJT) has also noted consistent mild AST elevation in patients with juvenile GM1 gangliosidosis.
In addition, compelling reports exist of patients with Alzheimers
disease [26], NPC and GM2g [27], and post-brain insult with elevated
AST levels in CSF [28]. Elevation of AST may be due to other concomitant medications (Table 1), such as carbamazepine and gabapentin,
which are commonly associated with hematological and hepatic

G.H.B. Maegawa et al. / Molecular Genetics and Metabolism 97 (2009) 284291

abnormalities [29,30]. One juvenile patient (HSC-104) developed a


persistent mild leucopenia, which was not clinically signicant. Leucopenia has also been described as a DRAE of carbamazepine, as well
of gabapentin, both concomitant medications were being taken by
this particular patient [30].
One of the limitations of the study was the small number of patients, a commonly encountered problem in studying patients with
rare disorders. As a result, the interpretations of the pharmacokinetic and safety results have to take into consideration the small
sample size.
The pharmacokinetics of miglustat in pediatric patients with
GM2g have been well described after both single- and multipledose administration, and were generally similar to results previously reported for adults with GD disease [18] and for healthy
adults [15]. Miglustat was generally safe and well tolerated,
although the DRAEs were different in the two cohorts studied. In
infantile patients, intermittent atulence/gasiness accompanied
by irritability were the most frequently observed and denitive
DRAEs. Whereas, among the juvenile patients, diarrhea, loose
stools, weight loss and vomiting were the most common DRAE.
Diarrhea and loose stools were classied as denitively related to
study drug, while neuropathy was classied as an unlikely DRAE
in the juvenile group, and more likely related to disease progression. Weight loss was observed only in juvenile patients and was
classied as probably drug-related. Laboratory abnormalities were
mostly mild and transient. Some persistent mild hematological and
hepatic changes were observed in some patients. Serum AST was
mild and consistently elevated in most patients in both cohorts.
Additional studies may clarify its origin.
We conclude that the pharmacokinetics of miglustat in children
are similar to those reported in adults, and that the drug is well tolerated with similar side effects as those reported in adults yet generally less severe with decreasing age.

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]
[18]
[19]

Acknowledgments
We thank Vivian Cruz, R.N., Lynn MacMillian, R.N., Margaret
Mackrell, R.N., and Barbara Chapman, R.N. who assisted in the clinical assessments. We are grateful for Dr. John Callahan and members of Clinical Biochemical Genetics Laboratory at the Hospital
for Sick Children who provided equipments for sample processing
and storage. Life for Luke Foundation supported the fellowship of
GHBM. We also acknowledge the nursing staff of the General Clinical Research Center at CNMC for sample collection and processing,
and the dedicated families of our patients. This study was supported in part by a grant from Actelion Pharmaceuticals, US and
by Grant 5-M01-RR-020359-04 from the National Center for Research Resources, NIH.
References
[1] R.H. Lachmann, F.M. Platt, Substrate reduction therapy for glycosphingolipid
storage disorders, Expert Opin. Investig. Drugs 10 (2001) 455466.
[2] T. Cox, R. Lachmann, C. Hollak, J. Aerts, S. van Weely, M. Hrebicek, F. Platt, T.
Butters, R. Dwek, C. Moyses, I. Gow, D. Elstein, A. Zimran, Novel oral treatment
of Gauchers disease with N-butyldeoxynojirimycin (OGT 918) to decrease
substrate biosynthesis, Lancet 355 (2000) 14811485.
[3] R. Heitner, D. Elstein, J. Aerts, S. Weely, A. Zimran, Low-dose Nbutyldeoxynojirimycin (OGT 918) for type I Gaucher disease, Blood Cells
Mol. Dis. 28 (2002) 127133.
[4] B. Bembi, F. Marchetti, V.I. Guerci, G. Ciana, R. Addobbati, D. Grasso, R. Barone, R.
Cariati, L. Fernandez-Guillen, T. Butters, M.G. Pittis, Substrate reduction therapy
in the infantile form of TaySachs disease, Neurology 66 (2006) 278280.
[5] J. Guitton, S. Coste, N. Guffon-Fouilhoux, S. Cohen, M. Manchon, M.
Guillaumont, Rapid quantication of miglustat in human plasma and
cerebrospinal uid by liquid chromatography coupled with tandem mass

[20]

[21]

[22]

[23]
[24]

[25]

[26]

[27]

[28]

[29]

[30]

291

spectrometry, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 877 (2009)


149154.
M.C. Patterson, D. Vecchio, H. Prady, L. Abel, J.E. Wraith, Miglustat for
treatment of NiemannPick C disease: a randomised controlled study, Lancet
Neurol. 6 (2007) 765772.
G.H. Maegawa, T. Stockley, M. Tropak, B. Banwell, S. Blaser, F. Kok, R. Giugliani,
D. Mahuran, J.T. Clarke, The natural history of juvenile or subacute GM2
gangliosidosis: 21 new cases and literature review of 134 previously reported,
Pediatrics 118 (2006) e1550e1562.
R. Gravel, M. Kaback, R. Proia, K. Sandhoff, K. Suzuki, K. Suzuki, The GM2
gangliosidosis, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The
Metabolic and Molecular Basis of Inherited Disease, McGraw-Hill, New York,
2001, pp. 38273876.
F.M. Platt, G.R. Neises, G. Reinkensmeier, M.J. Townsend, V.H. Perry, R.L. Proia,
B. Winchester, R.A. Dwek, T.D. Butters, Prevention of lysosomal storage in Tay
Sachs mice treated with N-butyldeoxynojirimycin, Science 276 (1997) 428
431.
M. Jeyakumar, T.D. Butters, M. Cortina-Borja, V. Hunnam, R.L. Proia, V.H. Perry,
R.A. Dwek, F.M. Platt, Delayed symptom onset and increased life expectancy in
Sandhoff disease mice treated with N-butyldeoxynojirimycin, Proc. Natl. Acad.
Sci. USA 96 (1999) 63886393.
J.F. Jacobs, M.A. Willemsen, J.J. Groot-Loonen, R.A. Wevers, P.M. Hoogerbrugge,
Allogeneic BMT followed by substrate reduction therapy in a child with
subacute TaySachs disease, Bone Marrow Transplant. 36 (2005) 925926.
B.E. Shapiro, G.M. Pastores, J. Gianutsos, C. Luzy, E.H. Kolodny, Miglustat in
late-onset TaySachs disease: a 12-month, randomized, controlled clinical
study with 24 months of extended treatment, Genet. Med., [Epub 5 May 2009
ahead of print].
Actelion, ZavescaTM 100 mg hard capsules (miglustat): summary of product
characteristics, Actelion Pharmaceuticas Inc. 2006, 2007. Available from:
<http://www.actelion.com>.
J.F. Commitee, British National Formulary for Children, British Medical
Association and Royal Pharmaceutical Society of Great Britain, London,
2005.
P.L.M. van Giersbergen, J. Dingemanse, Inuence of food intake on the
pharmacokinetics of miglustat, an inhibitor of glucosylceramide synthase, J.
Clin. Pharmacol. 47 (2007) 12771282.
NCI, Cancer Therapy Evaluation Program, Common Terminology Criteria for
Adverse Events, Version 3.0, DCTD, NCI, NIH, DHHS March 31, 2003, National
Cancer Institute, 2006.
R. Heitner, S. Arndt, J.B. Levin, Imiglucerase low-dose therapy for paediatric
Gaucher diseasea long-term cohort study, S. Afr. Med. J. 94 (2004) 647651.
P.L. McCormack, K.L. Goa, Miglustat, Drugs 63 (2003) 24272434.
I.L.O. Buxton, Pharmacokinetics and pharmacodynamics: the dymamics of
drug absorption, distribution, action and elimination, in: J.G. Hardmann, L.E.
Limbird, A.G. Gilman (Eds.), Goodman & Gilmans The Pharmacological Basis of
Therapeutics, McGraw-Hill Books, New York, 2006.
T.M. Cox, J.M. Aerts, G. Andria, M. Beck, N. Belmatoug, B. Bembi, R. Chertkoff, S.
Vom Dahl, D. Elstein, A. Erikson, M. Giralt, R. Heitner, C. Hollak, M. Hrebicek, S.
Lewis, A. Mehta, G.M. Pastores, A. Rolfs, M.C. Miranda, A. Zimran, The role of
the iminosugar N-butyldeoxynojirimycin (miglustat) in the management of
type I (non-neuronopathic) Gaucher disease: a position statement, J. Inherit.
Metab. Dis. 26 (2003) 513526.
G.M. Pastores, N.L. Barnett, E.H. Kolodny, An open-label, noncomparative study
of miglustat in type I Gaucher disease: efcacy and tolerability over 24 months
of treatment, Clin. Ther. 27 (2005) 12151227.
N.J. Weinreb, J.A. Barranger, J. Charrow, G.A. Grabowski, H.J. Mankin, P. Mistry,
Guidance on the use of miglustat for treating patients with type 1 Gaucher
disease, Am. J. Hematol. 80 (2005) 223229.
T.M. Cox, Substrate reduction therapy for lysosomal storage diseases, Acta
Paediatr. Suppl. 94 (2005) 6975.
E.P. Beltroy, B. Liu, J.M. Dietschy, S.D. Turley, Lysosomal unesteried
cholesterol content correlates with liver cell death in murine NiemannPick
type C disease, J. Lipid Res. 48 (2007) 869881.
H. Satoh, O. Yamato, T. Asano, M. Yonemura, T. Yamauchi, D. Hasegawa, H.
Orima, T. Arai, M. Yamasaki, Y. Maede, Cerebrospinal uid biomarkers showing
neurodegeneration in dogs with GM1 gangliosidosis: possible use for
assessment of a therapeutic regimen, Brain Res. 1133 (2007) 200208.
M. Riemenschneider, K. Buch, M. Schmolke, A. Kurz, W.G. Guder, Diagnosis of
Alzheimers disease with cerebrospinal uid tau protein and aspartate
aminotransferase, Lancet 350 (1997) 784.
S.M. Aronson, A. Saifer, G. Perle, B.W. Volk, Cerebrospinal uid enzymes in
central nervous system lipidoses (with particular reference to amaurotic
family idiocy), Proc. Soc. Exp. Biol. Med. 97 (1958) 331334.
E. Osuna, M.D. Perez-Carceles, A. Luna, D.J. Pounder, Efcacy of cerebro-spinal
uid biochemistry in the diagnosis of brain insult, Forensic Sci. Int. 52 (1992)
193198.
B. Parsons, L. Tive, S. Huang, Gabapentin: a pooled analysis of adverse events
from three clinical trials in patients with postherpetic neuralgia, Am. J. Geriatr.
Pharmacother. 2 (2004) 157162.
J.M. Pellock, Carbamazepine side effects in children and adults, Epilepsia 28
(Suppl. 3) (1987) S64S70.