Original Article CLINICAL

127
NEUROPHARMACOLOGY
Volume 31, Number 3
May - June 2008

A Study of Chorea After Tetrabenazine

Withdrawal in Patients With
Huntington Disease
Samuel Frank, MD,* William Ondo, MD,† Stanley Fahn, MD,‡
Christine Hunter, RN,† David Oakes, PhD,§ Sandra Plumb, BS,k
Frederick Marshall, MD,k Ira Shoulson, MD,k Shirley Eberly, MS,§
Francis Walker, MD,¶ Stewart Factor, DO,# Vicki Hunt, RN,¶
Aileen Shinaman, JD,k and Joseph Jankovic, MD†

Abstract patients.1 Currently, the extent of double-

Objective:
To assess tetrabenazine (TBZ) efficacy by evaluat-
ing the change in Huntington diseaseYassociated
chorea resulting from TBZ treatment withdrawal.
Methods:
Thirty patients treated in the long term were
randomized to 1 of 3 groups assigned to withdraw
from TBZ in a double-blind, staggered fashion
during a 5-day period.
Results:
The chorea scores of subjects withdrawn from TBZ
treatment increased by 5.3 units from days 1 to 3,
whereas the scores of the group with partial or no
withdrawal of TBZ treatment increased by 3.0
units (P = 0.0773). A post hoc analysis of the linear
trend was positive for reemergent chorea (P =
0.0486). No serious adverse events were reported
after abrupt withdrawal of TBZ treatment.
Conclusions:
The trend for reemergence of chorea in patients
with Huntington disease who were withdrawn
from TBZ treatment is consistent with the findings
from previous studies, thus showing the effective-
ness of TBZ in reducing chorea.
Key Words: chorea, Huntington disease,
tetrabenazine
(Clin Neuropharmacol 2008;31:127Y133)
blind studies of symptomatic treatments for
HD is limited, but neuroleptics, amantadine,
benzodiazepines, and other agents have
been used to reduce chorea.2 Several reports
have suggested that tetrabenazine (TBZ),
a monoamine-depleting drug, ameliorates
hyperkinetic movement disorders, includ-
ing chorea associated with HD.3Y10 In a
double-blind, placebo-controlled study of 84
patients with HD, TBZ was found to signifi-
cantly reduce chorea and provide a signifi-
cant benefit on ratings of clinical global
improvement.11 *Department of Neurology,
Boston University School of
Tetrabenazine selectively binds to the Medicine, Boston, MA;
†Department of Neurology,
centrally located vesicular monoamine trans-
Baylor College of Medicine,
porter (VMAT2) receptor and inhibits trans- Houston, TX; ‡Neurological
port of dopamine, norepinephrine, and Institute, Columbia University
College of Physicians and
serotonin into presynaptic vesicles.10 Vesic- Surgeons, New York,
ular monoamine transporter is located in the NY; Departments of
§Biostatistics and kNeurology,
presynaptic terminal, and inhibition of University of Rochester
VMAT2 results in selective depletion of School of Medicine, Rochester,
NY; ¶Department of
monoamines and serotonin.12Y14 The proc-
Neurology, Wake Forest
ess of VMAT binding and monoamine deple- University School of Medicine,
tion by TBZ is reversible, lasts hours, and is Winston-Salem, NC;
and #Department of
not modified by chronic treatment.15 Neurology, Emory University

H untington disease (HD) is an autosomal
dominant neurodegenerative disease
characterized by chorea, behavioral symp-
toms, and cognitive decline. Although there
is no established treatment to delay the
onset or slow down the progression of HD,
symptomatic treatment of chorea is avail-
able and may be beneficial for some
Chorea can subside at the final stages
of HD16Y19; thus, it may be useful to with-
draw patients periodically from the chorea-
suppressing agents to ensure that they still
need the drug. The primary objectives of this
study were to confirm the efficacy of TBZ by
demonstrating that chorea returns when the
drug is withdrawn from patients treated
School of Medicine, Atlanta, GA.
Supported by Prestwick
Pharmaceuticals.
Address correspondence and
reprint requests to Samuel
Frank, MD, 715 Albany St, C329,
Boston, MA 02118;
E-mail: samfrank@bu.edu
Copyright Ó 2008 by Lippincott
Williams & Wilkins

DOI: 10.1097/WNF.0b013e3180ca77ea

Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 CLINICAL Frank et al
128 NEUROPHARMACOLOGY
Volume 31, Number 3
May - June 2008
chronically with TBZ and to determine
the safety of short-term TBZ treatment
withdrawal.
MATERIALS AND METHODS
The subjects had manifest HD, as
confirmed by clinical diagnosis and an
expanded CAG repeat (n Q 37). The partici-
pants were on stable doses of TBZ for at least
2 months and had chorea that was deemed
clinically responsive to TBZ, as judged by the
investigator. They could not be taking con-
comitant dopamine-blocking or -depleting
agents, or monoamine oxidaseYinhibiting
agents. Antidepressants and benzodiazepines
were permitted, but only at stable doses.
All subjects were followed up at Baylor
College of Medicine Movement Disorder
Clinic and were withdrawn from TBZ treat-
ment in a double-blind fashion by replacing
TBZ with placebo tablets. Participants were
recruited during a 13-month period from
November 2003 to December 2004. Data
were maintained at the Clinical Trials Coor-
dination Center at the University of Roches-
ter, and the primary analysis was conducted
at the Department of Biostatistics, University
of Rochester. A 5-day washout protocol was
used on the basis of TBZ’s estimated mean of
5.5-hour half-life and of reports that chorea
usually relapses within 24 hours after inter-
ruption of treatment.8,20
The subjects were randomized using a
predetermined, computer-generated random-
ization scheme. Based on sequential random-
ization numbers, eligible subjects were
assigned to 1 of 3 parallel, unbalanced groups
during the 5-day study: (1) withdrawal group
(W group [withdraw TBZ treatment at day
1], 40%), (2) partial-withdrawal group (PW
group [withdraw TBZ treatment at day 3],
40%), and (3) no-withdrawal group (NW
group [remain on TBZ treatment], 20%)
(Fig. 1). Because of a consistent misinter-
pretation of the original protocol, the sub-
jects were generally evaluated before their
first morning dose of medication rather than
approximately 1 hour after it.
Ó 2008 Lippincott Williams & Wilkins
Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Baseline characteristics were compared
using Kruskal-Wallis tests or Fisher exact tests,
as appropriate. The primary analysis com-
pared the mean change from the baseline visit
to the day 3 evaluation in the Unified Hun-
tington Disease Rating Scale (UHDRS)21
chorea score in the W group, who were
withdrawn from TBZ treatment at day 1, with
the corresponding change in the combined
PW and NW groups, who were still on TBZ
at day 3. Analysis of covariance was used,
with the baseline chorea score as a covariate.
The original sample size calculation
called for a total of 45 subjects to be
enrolled, giving 84% power to detect a 3.1-
unit difference between the change score in
the W group (n = 18) and the change score
in the PW and the NW groups combined
(n = 27). This calculation was based on a
2-sided, level 0.05 test, assuming a pooled
standard deviation in the change of the total
maximum chorea score of 3.5 units.22 The
detectable effect corresponding to the
achieved sample size of 30 subjects was
calculated to be 4.0 units.
The intent-to-treat principle was used
for the primary analysis, with all randomized
participants analyzed according to their
originally assigned treatment group. Changes
from days 1 to 3 and from days 1 to 5 were
explored. A post hoc analysis to test for a
trend among the 3 groups was also per-
formed. The secondary outcome measure
was the change in total functional capacity
(TFC). The exploratory end points were the
changes in other UHDRS measures: total
motor score, cognitive assessment, behavio-
ral assessment, functional assessment, inde-
pendence scale, clinical summary scores,
Unified Parkinson Disease Rating Scale dys-
arthria score,23 selected individual items
from UHDRS (gait score, verbal fluency,
symbol digit modalities test, Stroop color
test, Stroop word test, Stroop interference),
and the Clinical Global Impression score.
Analyses of secondary and exploratory out-
come measures were conducted using the
same methodology as for the primary out-
come measure.
 Chorea After Tetrabenazine Withdrawal in HD Cases
FIGURE 1. Flow of participants.
The institutional review boards at Bay-
lor College of Medicine and the University of
Rochester approved this study. The partici-
pation of subjects in the study could be
discontinued anytime if they experienced
intolerable chorea after abrupt withdrawal
of TBZ treatment. After this study, the partici-
pants were offered the opportunity to par-
ticipate in a 1-year open-label study of TBZ.
RESULTS
Thirty subjects with HD were random-
ized into 3 parallel, unequal groups for this
5-day study: 12 in W group, 12 in PW group,
and 6 in NW group.
Baseline characteristics are shown
(Table 1). Ages ranged from 39 to 75 years
(median, 59 years). Baseline chorea scores
ranged from 2 to 25. The chorea scores in
Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
CLINICAL
NEUROPHARMACOLOGY
129
Volume 31, Number 3
May - June 2008
the 3 groups were not significantly different,
but the NW group had significantly worse
baseline total UHDRS motor scores (P =
0.0058) and significantly worse cognitive
scores. Generally, the NW group had more
advanced disease, as measured by duration
of disease and functional measures (TFC,
functional assessment, and independence
scale), although these differences did not
reach statistical significance. All 30 subjects
completed the protocol.
Mean chorea scores at the 3 time
points are shown in Figure 2, and the
unadjusted mean changes from baseline are
shown in Figure 3. The adjusted mean
chorea scores for W group increased by 5.3
units from days 1 to 3, whereas those in the
combined PW and NW groups increased by
3.0 units (P = 0.0773). Results were similar
when age (dichotomized at the median) was
 CLINICAL Frank et al
130 NEUROPHARMACOLOGY
Volume 31, Number 3
May - June 2008

TABLE 1. Baseline Characteristics

W Group PW Group NW Group
(n = 12) (n = 12) (n = 6)

Female, no. (%) 7 (58) 8 (67) 3 (50)

History of depression, no. (%) 7 (58) 8 (67) 4 (67)
Age, mean (SD), yrs 56.1 (9.7) 55.9 (8.5) 59.8 (14.2)
Duration of HD, mean (SD), yrs 10.1 (4.5) 8.9 (5.6) 11.4 (4.8)
Duration of TBZ treatment, mean (SD), yrs 1.2 (1.4) 0.9 (0.9) 1.5 (1.2)
Chronic dosage, mean (SD), mg 62.5 (38.1) 46.9 (18.6) 54.2 (24.6)
UHDRS subscales, mean (SD)
Chorea 9.4 (4.9) 9.1 (6.2) 11.2 (4.5)
Motor* 34.8 (9.7) 28.0 (12.0) 50.0 (8.7)
Symbol digit* 16.6 (5.6) 23.2 (9.4) 4.2 (3.4)
Stroop
Color naming* 40.5 (15.9) 41.4 (21.6) 18.3 (11.9)
Word reading 56.1 (26.9) 54.7 (26.9) 34.2 (25.3)
Interference* 22.6 (11.9) 23.5 (11.5) 4.6 (2.9)
Behavioral 15.3 (12.6) 17.1 (11.1) 22.2 (12.5)
Functional 15.9 (5.9) 17.3 (7.6) 12.7 (4.7)
Independence scale 72.5 (11.4) 75.0 (14.5) 65.0 (10.5)
TFC 6.3 (2.6) 7.6 (3.5) 5.0 (2.8)
*P G 0.05.

added to the model for the primary analysis;
a treatment by age interaction was not
significant (P = 0.2605). In view of the
baseline imbalance on total motor score, a
post hoc analysis of change in chorea score
adjusted for the baseline motor score (rather
than for the baseline chorea score) was
performed; the results were similar (P =
0.1202). The post hoc analysis of the linear
trend was positive for reemergent chorea
(P = 0.0486). There was no significant
change in the secondary efficacy end point
(TFC) or in the exploratory end points.
Twenty-nine of the 30 participants enrolled
in the open-label extension to the study.
During the study, 4 subjects were taking
concurrent dopaminergic antagonist med-
ications. Two subjects were taking agents

FIGURE 2. Mean (SE) UHDRS chorea scores.

Ó 2008 Lippincott Williams & Wilkins

Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Chorea After Tetrabenazine Withdrawal in HD Cases
FIGURE 3. Mean (SE) change in UHDRS chorea score.
specifically for chorea: one had been taking
haloperidol (dose, 0.25 mg) for 3 years, and
the other subject took fluphenazine (dose,
8 mg) for 2 years before entry into the study.
Another subject also took fluphenazine
(dose, 0.5 mg) for about 2.5 years for unclear
reasons. The fourth subject was taking que-
tiapine (dose, 100 mg) at bedtime for sleep
for about 6 months before enrollment. In the
intent-to-treat analysis, all subjects were
included. In the per-protocol analysis exclud-
ing subjects taking neuroleptics, comparison
of the days 1 to 3 adjusted change in chorea
scores still did not reach statistical signifi-
cance. The results were similar when the
primary analysis was adjusted for the use of
benzodiazepines. No increase in adverse
events was detected in the 4 subjects on
dopaminergic antagonist medications.
Adverse reactions to sudden treatment
withdrawal were generally mild and were
deemed either unrelated or unlikely to be
related to the study procedures by the treat-
ing investigator. Six subjects experienced 14
adverse events. The events occurred in 4
subjects in W group and 1 in each of the other
groups. Two subjects reported anxiety, and
2 subjects reported decreased appetite.
Diarrhea, dysphagia, hallucination, inflicted
injury (tongue laceration), insomnia, mood
swings, obsessive reaction, increased restless-
ness, sleep problems, and difficulty in swal-
lowing were each reported once. A single
Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
CLINICAL
NEUROPHARMACOLOGY
131
Volume 31, Number 3
May - June 2008
participant in the W group experienced 6
adverse events. No serious adverse events
were reported after abrupt withdrawal of TBZ
treatment. There were no early terminations
from the study.
DISCUSSION
Chorea is typically more prominent in
adult-onset, moderate early HD, whereas
dementia, bradykinesia, and dystonia become
more prominent as the disease progresses.16Y19
In this study, chorea reemerged in patients
who had been on TBZ treatment for as
long as 4.5 years. We were able to demon-
strate a significant trend in the reemergence
of chorea after sudden withdrawal of TBZ
treatment. However, there was no signifi-
cant difference in the primary outcome
measure of change in chorea between days
1 and 3 in the W group compared with the
combined PW and NW groups. Age did not
seem to play a role as to whether chorea
responded to TBZ treatment. The study also
demonstrated that sudden withdrawal of
TBZ treatment, up to a dose of 150 mg,
seems safe.
The design of this study was unique
because efficacy was evaluated by examining
the reemergence of chorea after withdraw-
ing TBZ treatment in a blinded, placebo-
controlled fashion rather than by observing
the improvement in choreic condition after
 CLINICAL Frank et al
132 NEUROPHARMACOLOGY
Volume 31, Number 3
May - June 2008
administering TBZ. However, because of the
limited availability of subjects already taking
TBZ for at least 2 months before random-
ization, enrollment was terminated on 30
subjects. The actual effect size of 2.3 units
was smaller than anticipated, in part, per-
haps, because of the systematic misunder-
standing of the protocol. In the original
study design, the participants were to be
examined before withdrawal of TBZ treat-
ment. In reality, at the time of the assess-
ments, the participants scheduled for TBZ
treatment withdrawal were already experi-
encing reemergence of chorea, reducing the
effect size. The lack of a significant differ-
ence in the worsening of UHDRS chorea
score between days 1 and 3 (primary out-
come measure) may have also been partially
caused by the nocebo effect or expectation
of sickness (placebo worsening of chorea) in
the NW group.24 Furthermore, although the
half-life of TBZ is 5.5 hours, there may be
pharmacodynamic or postsynaptic changes
with chronic administration that may not
have been detected in this short-term study.
Some evidence suggests that the half-life of
TBZ and its active metabolite varies mark-
edly in individual subjects.25
The significant trend for reemergence
of chorea in patients with HD who were
withdrawn from TBZ treatment is consis-
tent with the findings from our previous
studies, showing the effectiveness of TBZ in
reducing chorea. 11 Further studies are
required to characterize the magnitude and
the time frame for the reemergence
of chorea in HD and after TBZ treatment
withdrawal.
ACKNOWLEDGMENTS
The authors thank the patients who
participated in the study. They also thank
Prestwick Pharmaceuticals for contribution
to the study design.
REFERENCES
1. Frank SA, Marshall F, Plumb S, et al. CARE-HD
Investigators of the Huntington Study Group.
Ó 2008 Lippincott Williams & Wilkins
Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Functional decline due to chorea in
Huntington’s disease. Neurology 2004;62(suppl 5):
A204.
2. Jankovic J. Huntington’s disease. In: Noseworthy
J, ed. Neurological Therapeutics: Principles
and Practice. London, England: Martin Dunitz;
2006.
3. Dalby MA. Effect of tetrabenazine on extrapyramidal
movement disorders. Br Med J 1969;2(654):422Y423.
4. Jankovic J. Tetrabenazine in the treatment of
hyperkinetic movement disorders. Adv Neurol
1983;37:277Y289.
5. Jankovic J, Orman J. Tetrabenazine therapy of
dystonia, chorea, tics, and other dyskinesias.
Neurology 1988;38(3):391Y394.
6. Jankovic J, Beach J. Long-term effects of
tetrabenazine in hyperkinetic movement disorders.
Neurology 1997;48(2):358Y362.
7. Marsden CD. Drug treatment of diseases
characterized by abnormal movements. Proc R Soc
Med 1973;66(9):871Y873.
8. Paleacu D, Giladi N, Moore O, et al. Tetrabenazine
treatment in movement disorders. Clin
Neuropharmacol 2004;27(5):230Y233.
9. Fahn S. Long-term treatment of tardive dyskinesia
with presynaptically acting dopamine-depleting
agents. Adv Neurol 1983;37:267Y276.
10. Kenney C, Jankovic J. Tetrabenazine in the
treatment of hyperkinetic movement disorders.
Expert Rev Neurother 2006;6(1):7Y17.
11. Huntington Study Group. Tetrabenazine as
antichorea therapy in Huntington disease: a
randomized controlled trial. Neurology 2006;66(3):
366Y372.
12. Bagchi SP. Differential interactions of
phencyclidine with tetrabenazine and reserpine
affecting intraneuronal dopamine. Biochem
Pharmacol 1983;32(19):2851Y2856.
13. Butcher LL, Anden NE. Effects of apomorphine and
amphetamine on schedule-controlled behavior:
reversal of tetrabenazine suppression and
dopaminergic correlates. Eur J Pharmacol 1969;
6(3):255Y264.
14. Quinn GP, Shore PA, Brodie BB. Biochemical and
pharmacological studies of RO 1-9569
(tetrabenazine), a nonindole tranquilizing agent
with reserpine-like effects. J Pharmacol Exp Ther
1959;127:103Y109.
15. Scherman D, Jaudon P, Henry JP. Binding of a
tetrabenazine derivative to the monoamine
transporter of the chromaffin granule membrane
[In French]. C R Seances Acad Sci III 1981;293(4):
221Y224.
16. Young AB, Shoulson I, Penney JB, et al.
Huntington’s disease in Venezuela: neurologic
features and functional decline. Neurology 1986;
36(2):244Y249.
17. Bruyn G. Huntington’s chorea. Historical,
clinical and laboratory synopsis. In: Vincken P,
Bruyn G, eds. Diseases of the Basal Ganglia.
Handbook of Clinical Neurology. Amsterdam,
Netherlands: North Holland Publishing; 1968:
298Y378.
18. Hayden M. Huntington’s chorea. London, England:
Springer Verlag; 1981.
 Chorea After Tetrabenazine Withdrawal in HD Cases
19. Storey E, Beal MF. Neurochemical substrates of
rigidity and chorea in Huntington’s disease. Brain
1993;116(pt 5):1201Y1222.
20. Ondo WG, Tintner R, Thomas M, et al.
Tetrabenazine treatment for Huntington’s
diseaseYassociated chorea. Clin Neuropharmacol
2002;25(6):300Y302.
21. Huntington Study Group. Unified Huntington’s
Disease Rating Scale: reliability and consistency.
Mov Disord 1996;11(2):136Y142.
22. Huntington Study Group. Safety and tolerability of
the free-radical scavenger OPC-14117 in
Huntington’s disease. Neurology 1998;50(5):
1366Y1373.
Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
CLINICAL
NEUROPHARMACOLOGY
133
Volume 31, Number 3
May - June 2008
23. Fahn S, Elton RL. Members of the UPDRS
Developmental Committee. Unified Parkinson’s
Disease Rating Scale. In: Fahn S, Marsden CD,
Calne DB, et al, eds. Recent Development in
Parkinson’s Disease. Florham Park, NJ:
Macmillan Healthcare Information; 1987:
153Y163.
24. Hahn RA. The nocebo phenomenon: concept,
evidence, and implications for public health. Prev
Med 1997;26(5):607Y611.
25. Roberts MS, McLean S, Millingen KS, et al. The
pharmacokinetics of tetrabenazine and its hydroxy
metabolite in patients treated for involuntary
movement disorders. Eur J Clin Pharmacol 1986;
29(6):703Y708.