`Huntington disease
`A randomized controlled trial
`
`Huntington Study Group*
`
`Abstract—Background: Tetrabenazine (TBZ) selectively depletes central monoamines by reversibly binding to the type 2
`vesicular monoamine transporter. Open-label reports indicate TBZ is effective in treating chorea. Objective: To examine
`the safety, efficacy, and dose tolerability of TBZ for treating chorea in Huntington disease (HD). Methods: The authors
`randomized 84 ambulatory patients with HD to receive TBZ (n ⫽ 54) or placebo (n ⫽ 30) for 12 weeks. TBZ was increased
`over 7 weeks up to a maximum of 100 mg/day or until the desired antichoreic effect occurred or intolerable adverse effects
`supervened. The primary outcome was the change from baseline in the chorea score of the Unified Huntington’s Disease
`Rating Scale (UHDRS) Results: TBZ treatment resulted in a reduction of 5.0 units in chorea severity compared with a
`reduction of 1.5 units on placebo treatment (adjusted mean effect size ⫽ ⫺3.5 ⫾ 0.8 UHDRS units [mean ⫾ SE]; 95% CI:
`⫺5.2, ⫺1.9; p ⬍ 0.0001). There was also a significant benefit on ratings of clinical global improvement. There were five
`study withdrawals in the TBZ group and five serious adverse events (SAEs) in four subjects (drowning suicide, compli-
`cated fall, restlessness/suicidal ideation, and breast cancer) compared with one withdrawal and no SAEs in the placebo
`group. Conclusion: Tetrabenazine (TBZ), at adjusted dosages of up to 100 mg/day, effectively lessens chorea in ambulatory
`patients with Huntington disease. TBZ should be dosed individually based on ongoing assessment of possible adverse side
`effects.
`NEUROLOGY 2006;66:366–372
`
`Although there is no established treatment to delay
`the onset or forestall the progression of illness asso-
`ciated with Huntington disease (HD), symptomatic
`treatment of chorea may be beneficial in selected
`individuals.1,2 Neuroleptics reduce chorea but are as-
`sociated with extrapyramidal side effects. Several re-
`ports have suggested that tetrabenazine (TBZ)
`ameliorates hyperkinetic movement disorders,3-9 but
`controlled studies have been limited.10-12 TBZ is cur-
`rently marketed in nine countries including Canada,
`but it has been available in the United States only
`for research purposes.
`TBZ binds with high affinity (Kd ⫽ 2.4 nM) and
`selectivity to the CNS vesicular monoamine trans-
`porter (VMAT2),13-18 effectively depleting monoamines
`and serotonin (5HT) from nerve terminals by inhibiting
`their transport into presynaptic vesicles.14,19-21 TBZ
`depletes dopamine (IC50 ⫽ 0.4 mg/kg) preferentially
`
`Additional material related to this article can be found on the Neurology
`Web site. Go to www.neurology.org and scroll down the Table of Con-
`tents for the February 14 issue to find the title link for this article.
`
`over norepinephrine or 5HT (IC50 ⫽ 2 mg/kg), whereas
`reserpine, the only currently available monoamine de-
`pleter, is less selective and also binds to VMAT1 ex-
`pressed in the periphery.22-25 The highest binding
`density for TBZ is in the caudate nucleus, putamen,
`and nucleus accumbens, areas known to bear the brunt
`of pathology in HD.26,27 VMAT binding and mono-
`amine depletion by TBZ are reversible, last hours,
`and are not modified by chronic treatment,
`whereas those by reserpine are irreversible and
`last days to weeks.28,29 Unlike reserpine, TBZ has
`therefore not been associated with troublesome pe-
`ripheral side effects such as hypotension.
`We report the first multicenter, prospective,
`double-blind, placebo-controlled dose-finding study of
`TBZ for the treatment of chorea in HD.
`
`Methods. Subjects. Eligible subjects had HD as confirmed by
`the presence of a characteristic movement disorder (chorea), a
`family history, and an expanded CAG repeat (n ⱖ 37). All partici-
`pants were required to be independently ambulatory, to have a
`screening total functional capacity (TFC) of ⬎5,30 and a total max-
`imal chorea of ⱖ10 (sum of the maximal chorea scores for facial,
`
`*See the Appendix for a complete listing of authors.
`From the Clinical Trials Coordination Center, Department of Neurology, University of Rochester School of Medicine and Dentistry, NY.
`Disclosure: This study was funded by a grant from Prestwick Pharmaceuticals, Inc., to the University of Rochester and in turn through subcontracts to the
`participating research sites. The Huntington Study Group (HSG) is a nonprofit consortium of Huntington’s disease investigators (http://www.huntington-
`study-group.org/). None of the HSG investigators or staff had equity interests with Prestwick Pharmaceuticals, Inc. Dr. Fahn received consulting fees of less
`than $10,000. Dr. Marshall presented data at national meetings for which he received travel reimbursement from Prestwick Pharmaceuticals, Inc. Dr.
`Clarence-Smith, Dr. O’Brien, and Ms. Wilson are employees of Prestwick Pharmaceuticals, Inc. The HSG Coordination and Biostatistics Centers at the
`University of Rochester independently compiled and analyzed the data for this study.
`Received April 4, 2005. Accepted in final form October 18, 2005.
`Address correspondence and reprint requests to Dr. F.J. Marshall, Clinical Trials Coordination Center, Department of Neurology, University of Rochester
`School of Medicine and Dentistry, 1359 Mt. Hope Ave., Suite 223, Rochester, NY 14620; e-mail: fred.marshall@ctcc.rochester.edu
`
`366 Copyright © 2006 by AAN Enterprises, Inc.
`
`Apotex Ex. 1010
`
`Apotex v. Auspex
`IPR2021-01507
`
`
`
`buccal–oral–lingual, trunkal, and each extremity from the motor
`subscale of the Unified Huntington’s Disease Rating Scale
`[UHDRS]).31 Patients were excluded if they had disabling depres-
`sion, dysphagia, or dysarthria. Eligible participants could not
`have been treated in the past with TBZ or currently with
`dopamine-depleting medications, dopamine D2 receptor blockers,
`selective or nonselective monoamine oxidase inhibitors, levodopa,
`dopamine agonists, amantadine, or memantine. Patients previ-
`ously treated with dopamine D2 receptor blockers were enrolled,
`provided they had been off these medications for at least 4 weeks.
`Patients were permitted to take existing antidepressant or benzo-
`diazepine medication if receiving stable dosages for at least 8
`weeks prior to the randomization visit. Subjects needed to be
`accompanied by a caregiver.
`Study design. This
`randomized, double-blind, placebo-
`controlled study was carried out at 16 Huntington Study Group
`(HSG) sites in the United States. The primary purpose was to
`study the efficacy, dosing, tolerability, and safety of TBZ in two
`parallel groups of HD subjects with clinically manifest chorea,
`allocated 2:1 to receive TBZ or placebo. The study protocol re-
`ceived institutional review board approval at each participating
`center prior to subject enrollment.
`A data and safety monitoring committee consisting of two phy-
`sicians experienced in clinical research and a biostatistician, who
`were not otherwise involved in the conduct of the study, convened
`on three occasions during the course of the trial to assess safety,
`including unblinded analysis of accrued adverse events and labo-
`ratory tests.
`Study procedures. Consenting research participants who sat-
`isfied the eligibility criteria were randomized (2:1) to receive ei-
`ther TBZ or placebo for 12 weeks. Treatment assignment was
`concealed from subjects and investigators, and randomization was
`performed through a computerized module developed in the De-
`partment of Biostatistics at the University of Rochester. TBZ was
`formulated in 12.5-mg tablets and was identical in appearance to
`matching placebo tablets. During the first 7 weeks of the study,
`dosage was titrated in blinded fashion by providing 1 tablet on the
`first day, then 1 tablet twice daily for the remainder of the first
`week. Subsequently, the number of tablets was increased by 1 per
`week up to 8 tablets per day in three divided doses or until a
`desired antichoreic effect was achieved or intolerable adverse ef-
`fects occurred. To reduce intolerable side effects, the number of
`tablets was reduced to the participant’s previously well-tolerated
`level or lower if necessary. Participants were permitted one sus-
`pension of study drug for up to 7 days. By the end of the first 7
`weeks of the study, participants were on their “best dose,” and
`during the last 5 weeks of the study, the dosage remained con-
`stant unless reduced because of intolerable adverse effects. After
`12 weeks, study drugs were withdrawn, and subjects returned for
`a follow-up visit 1 week later.
`Assessments. Subjects were examined at baseline and at the
`end of weeks 1, 3, 5, 7, 9, 12, and 13. At each examination, the
`investigator rated the UHDRS total maximal chorea score. The
`Clinical Global Impression (CGI) was done at weeks 1, 3, 5, 7, 9,
`and 12. A full UHDRS, including motor, cognitive, behavioral, and
`functional components, was completed at baseline, at the end of
`the titration phase, and at the end of the maintenance phase.31
`Complete blood count and general chemistry profiles were ob-
`tained at screening and at the end of week 12.
`Participants were also evaluated for adverse events, parkin-
`sonism subscore of the UHDRS (sum of finger taps, pronate/
`supinate hands, rigidity arms, bradykinesia–body, gait, tandem
`walking, and retropulsion pull test), akathisia as measured by the
`Barnes Akathisia Scale,32 speech and swallowing as measured by
`Unified Parkinson’s Disease Rating Scale Part II, item 5 (UPDRS
`speech) and item 7 (UPDRS swallowing),33 depression as mea-
`sured by the 17-item Hamilton Depression Scale (HAM-D),34
`sleepiness as measured by the Epworth Sleepiness Scale (ESS),35
`vital signs (systolic and diastolic blood pressure, radial artery
`pulse rate) while sitting, and a 12-lead EKG. We also piloted a
`new instrument, the Functional Impact Scale (FIS), to assess the
`degree of difficulty with bathing, dressing, feeding, social isola-
`tion, and toileting. FIS information was obtained from the accom-
`panying caregiver, and each item was graded on a scale of 0 to 3
`(see appendix E-1 on the Neurology Web site; go
`to
`www.neurology.org).
`An amendment to the initial protocol, adding a videotape to
`
`document subjects’ chorea at the end of the maintenance phase
`(week 12) and after withdrawal (week 13), was instituted during
`the course of the study. An independent movement disorder ex-
`pert who was blinded to subjects’ treatment assignments, as well
`as to whether a given tape had been done at the conclusion of the
`maintenance phase or the conclusion of the washout phase, rated
`each tape as to chorea severity using the UHDRS and overall
`clinical improvement using the CGI Improvement Scale.
`Statistical analysis. The primary prespecified efficacy out-
`come measure was the difference between the baseline total max-
`imal chorea score and the average of the score at week 9
`(midmaintenance phase) and week 12 (end of maintenance phase).
`Analysis of covariance (ANCOVA) was used to assess the treat-
`ment effect. The model included the site and baseline score. The
`treatment-by-site interaction was not significant and was there-
`fore not included in the final model. Primary analyses were based
`on intention to treat. If either the week 9 or the week 12 data
`were missing for a given subject, the missing data were imputed
`as the one available score. If both week 9 and week 12 data were
`missing, the subject’s last available score (after baseline) was car-
`ried forward.
`Power calculations were based on the results obtained in the
`subgroup of patients whose baseline total maximal chorea score
`was at least 10 in a previous 12-week HSG trial. Given a 2:1
`randomization (TBZ/placebo), an estimated dropout rate of 15%,
`and an estimated baseline chorea score of 14.5 ⫾ 3.5 (mean ⫾ SD),
`a total sample size of 72 subjects gave ⬎80% power to detect an
`effect size of at least 2.7-unit change in the total maximal chorea
`score. The study sample provided a probability of ⬎90% for detect-
`ing an adverse event that occurred with a frequency of 10% or
`more in the population from which the participants were drawn
`and a power of ⬎99% for detecting an absolute difference in toler-
`ability of 50% between the placebo group and the TBZ group.
`Because of the large number of potential outcome measures,
`the analysis plan prespecified that four secondary endpoints (CGI
`Global Improvement score, change in total motor score, change in
`functional checklist, change in gait score) be analyzed in descend-
`ing hierarchy, with definitive analyses ceasing when the signifi-
`cance level reached ␣ ⱖ 0.05. The CGI Global Improvement score
`was analyzed using only data from week 12, as prespecified in the
`analysis plan. ANCOVA procedures comparable with those in the
`primary efficacy analysis were used for the change from baseline
`in the other measures. Exploratory analyses of other outcome
`measures were undertaken after the hierarchical analyses
`reached a stopping point. Baseline characteristics were compared
`using t tests or 2 tests, as appropriate. Group comparisons on
`tolerability measures and adverse events were made with
`continuity-corrected 2 tests or F tests.
`
`Results. Baseline comparability.
`Figure 1 outlines the
`flow of participants in the study. Ninety-one individuals
`were screened and 84 eligible subjects were randomized
`between July and December 2003. Table E-1 lists the base-
`line demographic and outcome variable scores for all par-
`ticipants. The groups were comparable with regard to age,
`CAG repeat length, gender, duration of illness, and propor-
`tion with a history of depression. Subjects randomized to
`TBZ scored worse at baseline on the Symbol Digit Test of
`the UHDRS cognitive battery (p ⫽ 0.018), as well as on the
`pilot FIS (p ⫽ 0.035). Although the other individual
`UHDRS measures were not statistically different across
`treatment groups, subjects randomized to TBZ tended to
`score somewhat more poorly at baseline on most UHDRS
`measures of severity with the notable exception of chorea
`score (see table E-1). There was a strong inverse correla-
`tion between CAG repeat length and the age at symptom
`onset in our study sample (r ⫽ ⫺0.63, p ⬍ 0.0001), as
`observed in other studies of HD patients.36,37
`Outcome summaries. Table 1 shows the adjusted mean
`changes in treatment-related outcome measures from
`baseline to the prespecified study endpoint (average of
`week 9 and week 12 scores), including the adjusted treat-
`February (1 of 2) 2006 NEUROLOGY 66 367
`
`Apotex Ex. 1010
`
`
`
`Primary efficacy outcome: Impact of TBZ on cho-
`rea. Although there was a mild reduction in total maxi-
`mal chorea in the placebo group between baseline and the
`average of the week 9 and week 12 assessments (⫺1.5 ⫾
`0.7 UHDRS units; adjusted mean ⫾ SE), the impact of
`TBZ on chorea severity was larger: ⫺5.0 ⫾ 0.5 UHDRS
`units (p ⬍ 0.0001). The adjusted effect size of ⫺3.5 UH-
`DRS units (95% CI: ⫺5.2, ⫺1.9), represents a 23.5% aver-
`age reduction in baseline chorea severity due to TBZ
`(figure 2). Whereas only 20% of placebo subjects had a
`reduction in chorea of at least 3 UHDRS units, 69% of
`TBZ-treated subjects had a reduction of at least this mag-
`nitude (adjusted odds ratio ⫽ 9.9; 95% CI: 3.2, 29.9; p ⬍
`0.0001). The TBZ-related reduction in chorea was not re-
`lated to age, gender, trinucleotide repeat length, gender of
`affected parent, baseline CGI of severity, or baseline cho-
`rea score.
`Secondary efficacy outcomes. TBZ was superior to pla-
`cebo on the CGI Global Improvement Scale, with an ad-
`justed effect size (improvement) of ⫺0.7 CGI unit (95% CI:
`⫺1.3, ⫺0.2) on this 7-point scale. Figure 3 shows the distri-
`bution of CGI Global Improvement scores at week 12 by
`treatment group. Twenty-four percent of subjects in the
`placebo group achieved a CGI Global Improvement score of
`ⱕ3 (corresponding to at least minimal global improve-
`ment) compared with 69% of TBZ subjects (p ⫽ 0.0001).
`Only two participants receiving placebo (6.9%) had more
`than minimal global improvement, whereas 23 TBZ partic-
`
`Figure 1. Flow of study subjects.
`
`ment effect with 95% CIs, the effect sizes as a function of
`the average baseline scores, and the effect sizes as a func-
`tion of the total scale range for each variable.
`
`Table 1 Change in outcome variables (adjusted mean ⫾ SE) from baseline to average of week 9 and week 12 scores
`(ITT/LOCF analyses)
`
`Direction of
`favorable
`change
`
`Placebo,
`n ⫽ 30
`
`TBZ,
`n ⫽ 54
`
`p Value
`⬍ 0.05
`
`Adjusted
`mean TE,
`scale units
`
`TE 95%
`CI, scale
`units
`
`TE,
`% baseline
`mean score
`
`TE,
`% scale
`range
`
`Primary outcome variable
`⌬ UHDRS tot max. chorea
`Secondary outcome variables
`CGI Global Improvement‡
`⌬ UHDRS total motor
`Exploratory outcome variables
`⌬ UHDRS functional checklist
`⌬ 17-item HAM-D
`⌬ Epworth Sleepiness
`⌬ Stroop test
`Word reading
`Color naming
`Interference
`
`⫺
`
`⫺
`
`⫹
`
`⫺
`
`⫺
`
`⫹
`
`⫹
`
`⫹
`
`⫺1.5 ⫾ 0.7 ⫺5.0 ⫾ 0.5
`
`0.0001*
`
`⫺3.5†
`
`⫺5.2, ⫺1.9
`
`⫺23.5
`
`⫺12.5
`
`3.0 ⫾ 0.2
`3.7 ⫾ 0.2
`⫺3.5 ⫾ 1.5 ⫺6.8 ⫾ 1.1
`
`0.007*
`
`0.4 ⫾ 0.4 ⫺0.8 ⫾ 0.3
`⫺2.4 ⫾ 0.4 ⫺0.7 ⫾ 0.3
`⫺0.3 ⫾ 0.6
`1.5 ⫾ 0.5
`
`0.02§
`0.003§
`0.02§
`
`1.8 ⫾ 2.1 ⫺4.8 ⫾ 1.5
`1.3 ⫾ 1.7 ⫺1.7 ⫾ 1.2
`1.5 ⫾ 1.2 ⫺1.5 ⫾ 0.9
`
`0.01§
`
`⫺0.7
`⫺3.3
`
`⫺1.2
`1.6
`1.8
`
`⫺6.6
`⫺2.9
`⫺3.0
`
`⫺1.3, ⫺0.2
`⫺7.0, 0.3
`
`⫺2.2, ⫺0.2
`0.6, 2.7
`0.3, 3.4
`
`⫺11.8, ⫺1.5
`⫺7.3, 1.4
`⫺6.0, 0.0
`
`⫺7.1
`
`⫺6.3
`33.8
`47.7
`
`⫺12.1
`⫺6.6
`⫺12.9
`
`⫺10.0
`⫺2.7
`
`⫺4.8
`3.1
`7.5
`
`—
`—
`—
`
`Per the prespecified hierarchical analysis plan (see text), only changes in total maximal chorea and CGI Global Improvement scores
`achieved definitive significance.
`
`* Favors TBZ over placebo.
`† The study was powered to detect an effect size of 2.7 UHDRS units.
`‡ CGI Global Improvement analysis based on week 12 rating per analysis plan. For CGI Global Improvement, a score of ⬍4 ⫽ improve-
`ment, 4 ⫽ no change, ⬎4 ⫽ worsening.
`§ Favors placebo over TBZ.
`
`ITT ⫽ intention-to-treat analysis; LOCF ⫽ last observation carried forward analysis; TBZ ⫽ tetrabenazine; TE ⫽ treatment effect (TBZ
`compared with placebo); UHDRS ⫽ Unified Huntington’s Disease Rating Scale; HAM-D ⫽ Hamilton Depression Scale; CGI ⫽ Clinical
`Global Impression; UPDRS ⫽ Unified Parkinson’s Disease Rating Scale.
`
`368 NEUROLOGY 66 February (1 of 2) 2006
`
`Apotex Ex. 1010
`
`
`
`Figure 2. Mean change from baseline
`in Unified Huntington’s Disease Rating
`Scale total maximal chorea score by
`treatment group (last observation car-
`ried forward [LOCF] except for week
`13). Change from baseline to week 12
`favors tetrabenazine (p ⫽ 0.0001; anal-
`ysis of covariance, intention to treat,
`LOCF, side panel). There was blinded
`washout of study drug after week 12.
`Scale range: 0 to 28.
`
`ipants (45.1%) were more than minimally improved (p ⫽
`0.0004). There was a correlation between improvement in
`chorea and the CGI Global Improvement score at week 12
`(r ⫽ 0.58, p ⬍ 0.0001, n ⫽ 80). A similar correlation was
`found between improvement in total motor score and CGI
`Global Improvement score at week 12 (r ⫽ 0.41, p ⫽
`0.0001).
`Although there was a trend toward an improvement,
`the impact of TBZ on the UHDRS total motor score did not
`reach significance (adjusted mean treatment effect ⫽ ⫺3.3
`UHDRS units; 95% CI: ⫺7.0, 0.3; p ⫽ 0.08). Per the se-
`quential hierarchy of analyses prespecified in the study
`protocol, all further analyses of secondary outcomes were
`considered exploratory rather than definitive once this re-
`sult was established.
`Exploratory outcome measures. There was an adverse
`impact of TBZ on the UHDRS Functional Checklist, with
`the placebo group improving by 0.4 unit on this 25-point
`scale and the TBZ worsening by 0.8 unit (p ⫽ 0.02). There
`was no impact of TBZ on UHDRS gait assessment,
`UHDRS parkinsonism score, or any of the other explor-
`atory outcome measures (TFC, FIS, Independence Scale).
`There was a small but significant correlation between
`worsening UHDRS Functional Checklist scores and wors-
`ening UHDRS parkinsonism scores (r ⫽ 0.24; p ⫽ 0.027).
`Subjects in the TBZ group reported more sleepiness on the
`ESS (adjusted mean effect size ⫽ 1.8 ESS units; 95% CI:
`0.3, 3.4; p ⫽ 0.02). There was no significant impact of TBZ
`on the Barnes Akathisia Scale, UHDRS Behavioral Assess-
`ment (calculated as the sum of all items), UPDRS swallow-
`
`ing, or UPDRS speech items. TBZ had an adverse impact
`on Stroop word reading, but not on other UHDRS
`measures of cognitive function. Subjects in both the pla-
`cebo and the TBZ groups improved on the 17-item HAM-D
`over the course of the trial, but those in the placebo group
`did so to a slightly greater degree, with a relatively smaller
`improvement seen in the TBZ group (adjusted mean effect
`size ⫽ 1.6 HAM-D units; 95% CI: 0.6, 2.7; p ⫽ 0.003). As
`the mean HAM-D score across all subjects at baseline was
`only 4.7, the net TBZ effect was clinically insignificant in
`the context of the threshold score of ⬎12 for a diagnosis of
`depression. There was, however, a small but significant
`correlation between worsening HAM-D scores and worsen-
`ing UHDRS Functional Checklist scores (r ⫽ 0.30;
`p ⫽ 0.006).
`Washout period. There were no differences between
`TBZ and placebo at the conclusion of the washout phase
`(week 13), compared with baseline, with regard to motor,
`cognitive, behavioral, or global measures of illness sever-
`ity. Chorea in subjects on TBZ worsened more than pla-
`cebo subjects in the period following withdrawal of
`medication at week 12 (adjusted effect size ⫽ 4.4 UHDRS
`units; 95% CI: 2.8, 6.0; p ⬍ 0.0001). These results were
`confirmed by analysis of chorea ratings and CGI ratings
`made by the independent movement disorder expert on a
`subset of 23 subjects for whom videotapes at weeks 12 and
`13 were available. Video chorea worsening attributable to
`the washout from TBZ vs placebo was estimated at 4.08
`UHDRS units (ANCOVA, 95% CI: 0.55, 7.62; p ⫽ 0.03).
`Seventy-one percent (10/14) of the TBZ patients were at
`
`Figure 3. Distribution of Clinical
`Global Impression Global Improvement
`ratings at week 12 (end of active treat-
`ment phase) by treatment group. Scores
`are as follows: 1 ⫽ very much im-
`proved, 2 ⫽ much improved, 3 ⫽ mini-
`mally improved, 4 ⫽ no change, 5 ⫽
`minimally worse, 6 ⫽ much worse, 7 ⫽
`very much worse. There is a shift of the
`curve to the left (improvement) in the
`tetrabenazine group (p ⫽ 0.0001 for
`proportion achieving score of ⱕ3; 2
`intention to treat).
`
`February (1 of 2) 2006 NEUROLOGY 66 369
`
`Apotex Ex. 1010
`
`
`
`Table 2 Tolerability analyses
`
`Variable
`
`Subjects withdrawn (see text)
`Subjects experiencing at least one SAE (see text)
`New AEs per subject, mean ⫾ SD
`All
`Excluding mild
`Subjects reporting AEs
`All
`Excluding mild
`Subjects with week 12 reduced dosage due to intolerability
`
`TBZ ⫽ tetrabenazine; SAE ⫽ serious adverse event.
`
`least minimally worse on the CGI after withdrawal of drug
`compared with 22% (2/9) of the placebo patients (Coch-
`ran—Armitage trend test, p ⫽ 0.01). There was strong
`correlation between the independent rater’s assessment of
`chorea severity by video and investigators’ reports of cho-
`rea severity in the clinic at both week 12 (r ⫽ 0.76; p ⬍
`0.0001) and at week 13 (r ⫽ 0.68; p ⫽ 0.0004).
`Tolerability and serious adverse events. Seventy-eight
`(93%) of the 84 subjects completed the full 13 weeks of the
`study. One subject in the placebo group was lost to follow-
`up. Five withdrawals occurred in the TBZ group: four due
`to serious adverse events (SAEs), and one due to akathisia.
`Four participants receiving TBZ experienced five SAEs.
`There was one death by drowning due to suicide. Neither
`the investigator nor the subject’s caregiver had detected
`signs of depression at a study visit conducted 2 weeks prior
`to the event, and no abnormality was detected on the
`HAM-D at that time. There was a premature disclosure of
`treatment assignment to the site investigator consequent
`to this SAE. One TBZ-treated subject had an intracerebral
`hemorrhage consequent to a fall and subsequently re-
`turned to baseline level of function. One participant was
`hospitalized for restlessness that resolved within 48 hours
`of dosage reduction and treatment with clonazepam, but 2
`weeks later developed depressive symptoms, irritability,
`and suicidal ideation when motor symptoms worsened.
`The subject was continued on TBZ, rehospitalized, and
`treated with mirtazapine, with resolution of suicidal ide-
`ation within a day of rehospitalization. One subject had
`been aware of a breast lump prior to screening but did not
`bring it to the investigator’s attention until after she was
`enrolled; she was diagnosed with breast cancer during the
`study. There were no SAEs in the placebo group.
`All AEs. Table 2 summarizes the number of with-
`drawals, dosage reductions due to intolerability, subjects
`experiencing at least one AE, and number of AEs per sub-
`ject. Twenty-one (70%) of placebo and 49 (91%) of TBZ
`participants experienced an AE (World Health Organiza-
`tion preferred coding; p ⫽ 0.01). Table E-2 lists AEs re-
`ported in four or more participants (approximately 5% of
`the total study population). By the conclusion of the main-
`tenance phase, when subjects were presumably on optimal
`dosage, there were no significant differences between TBZ
`and placebo with regard to specific AEs that had not been
`reported at baseline (coding based on the AE log using
`World Health Organization preferred term). Among sub-
`
`370 NEUROLOGY 66 February (1 of 2) 2006
`
`Placebo, n (%), n ⫽ 30
`
`TBZ, n (%), n ⫽ 54
`
`p Value
`
`1 (3.3)
`0 (0)
`
`1.5 ⫾ 1.8
`0.6 ⫾ 0.9
`
`21 (70.0)
`10 (33.3)
`1 (3.3)
`
`5 (9.3)
`4 (7.4)
`
`3.8 ⫾ 3.1
`1.9 ⫾ 2.0
`
`49 (90.7)
`37 (68.5)
`24 (44.4)
`
`NS
`NS
`
`0.0005
`0.0007
`
`0.01
`0.002
`⬍0.0001
`
`jects completing the study, the most common AE at week
`12 was fatigue, reported by seven subjects on TBZ (14.3%)
`and two subjects on placebo (6.9%).
`Dosage adjustments. Figure E-1 on the Neurology Web
`site gives the distribution of dosages achieved by each
`group at the end of the titration and maintenance phases.
`Twenty-seven (55%) of the 49 TBZ subjects and 4 (14%) of
`the 29 placebo subjects completing the study were taking
`less than the maximum allowed dosage at the end of the
`maintenance phase (week 12). Of these, one TBZ and two
`placebo subjects had early washouts during the mainte-
`nance phase (unrelated to adverse events), whereas two
`TBZ and one placebo subject achieved desirable anticho-
`reic effect at less than maximal dosage. Dose-limiting
`symptoms (based on the dosage adjustment log) in the
`TBZ-treated group included sedation in 13 (27%), akathi-
`sia in 4 (8%), parkinsonism in 2 (4%), depression as de-
`scription of mood rather than a formal diagnosis in 2 (4%),
`and other in 3 (6%). One placebo-treated subject (3%) re-
`ported light-headedness as a dose-limiting symptom.
`Vital signs and laboratory results. There were no sig-
`nificant treatment differences in blood pressure or weight
`between baseline and the end of the maintenance phase.
`There was an increase in pulse in the TBZ group (5.9 ⫾ 2.7
`beats/min, adjusted mean effect ⫾ SE; p ⫽ 0.03). There
`were no clinically relevant EKG changes during the course
`of the study. All participants had laboratory values less
`than grade 3 by National Cancer Institute guidelines at
`screening, with the exception of lipid profiles. Three sub-
`jects receiving TBZ developed alanine aminotransferase
`(ALT) increases that were not associated with elevated
`bilirubin values or clinical symptoms. Of these three pa-
`tients, two were classified as grade 2 and resolved with
`continued therapy in an open extension trial. The patient
`who developed a grade 3 increase had an abnormal value
`at baseline and acknowledged binge drinking at the time of
`the increase; this patient’s ALT normalized on withdrawal
`of therapy after which he tolerated retreatment at a lower
`dose in an open extension trial. There were no other clini-
`cally meaningful differences between the groups with re-
`gard to changes in laboratory values during the course of
`the study, and the groups did not differ statistically with
`regard to the occurrence of ALT elevation.
`Medications. The most common concomitant medica-
`tions at study entry included antidepressants (60%) and
`
`Apotex Ex. 1010
`
`
`
`benzodiazepines (17%). There was no difference between
`groups in the use of these medications.
`
`Discussion. TBZ significantly reduced chorea bur-
`den, improved global outcome, and was generally
`safe and well tolerated in this 12-week study. The
`antichoreic impact of TBZ was clinically meaningful
`and statistically significant, with benefit in a global
`measure of clinical outcome (CGI Global Improve-
`ment Scale) that correlated well with the degree of
`chorea improvement. The 3.5-unit (23.5%) reduction
`in chorea severity seen in the TBZ group, compared
`with placebo, contrasts favorably with that seen in
`recently reported trials of remacemide (0.9 unit,
`9.8%)36 and riluzole (2.2 units, 18.0%).37 Amantadine
`was previously reported to improve chorea by 36%,1
`but a more recent double-blind controlled study
`showed no benefit of amantadine on chorea.38 A post-
`hoc analysis of chorea response by dosage in our
`study indicates that subjects whose TBZ dosage was
`50 mg/day or less had a greater improvement in cho-
`rea at week 12 (adjusted effect size compared with
`placebo: ⫺6.24; 95% CI: ⫺8.56, ⫺3.93) than did those
`receiving more than 50 mg/day (⫺3.65; 95% CI:
`⫺5.70, ⫺1.60). Differences in susceptibility to TBZ
`due to hepatic processing, avidity of VMAT-2, or
`other unknown factors may account for this finding.
`Whereas chorea is often treated with standard or
`atypical neuroleptics, there are no controlled clinical
`data available to judge the degree of chorea reduc-
`tion achieved with these agents. Adverse effects of
`neuroleptics as antichoreic treatment include par-
`kinsonism, decreased balance, akathisia, neuroleptic
`malignant syndrome, acute dystonic reactions, tar-
`dive dyskinesia, blunting of affect, and generalized
`apathy. Based on this study and the pharmacology of
`TBZ, the typical extrapyramidal effects seen with
`neuroleptics appear to be less common. Because our
`study was only 12 weeks long, we cannot draw infer-
`ences about adverse effects due to TBZ that may
`emerge after more prolonged exposure.
`The rapid recurrence of chorea during the wash-
`out phase is likely due to the relatively short half-life
`of the drug (approximately 4 hours), but a placebo
`effect cannot be completely excluded. A staggered
`washout study would be informative.
`An unexplained outcome of this study is the dis-
`cordant improvement of TBZ-treated patients on the
`CGI Global Improvement Scale in comparison with
`the slight worsening on the exploratory analysis of
`the UHDRS Functional Checklist. No adverse im-
`pact of TBZ on other measures of function (Indepen-
`dence Scale, FIS) was found. One explanation of this
`finding is that the UHDRS Functional Checklist
`failed to incorporate items of functional significance
`relevant to potentially beneficial effects of TBZ, such
`as the ability to carry out specific activities (e.g.,
`eating or dressing). We noted a significant correla-
`tion between deterioration on the HAM-D or on the
`UPDRS Parkinsonism Scale and worsening UHDRS
`Functional Checklist scores. The fact that improve-
`
`ment in chorea was significantly correlated to im-
`provement in the CGI Global Improvement Scale
`underscores the contribution that chorea makes to
`the overall clinical burden of HD and validates the
`use of the UHDRS chorea score as a clinical outcome
`measure.
`TBZ was associated with a significant increase in
`reports of drowsiness and insomnia, though only four
`subjects reported both symptoms. Other dose-
`limiting AEs included depressed mood (n ⫽ 2), par-
`kinsonism (n ⫽ 2), and akathisia (n ⫽ 4). These AEs
`generally resolved with dosage adjustment. There
`were no significant differences between TBZ and pla-
`cebo in the number of AEs present on optimized dos-
`ages at week 12. We detected no adverse impact at
`week 12 on the Barnes Akathisia Rating Scale or on
`measures of parkinsonism, dysphagia, or dysarthria.
`There was no evidence of TBZ-related depression as
`determined by the HAM-D,17 though patients in the
`placebo group had a greater improvement in this
`index of mood than did those in the TBZ group over
`the 12-week study.
`One subject in the TBZ group committed suicide,
`despite scoring within the normal range on the
`HAM-D 2 weeks prior to the event. Clinicians should
`be cautious about the heightened risk of suicide in
`HD regardless of depression indexes or use of mono-
`amine depleters, as overall completed suicide rates
`are estimated at 7.3% and up to 25% of patients may
`attempt suicide at some point in the illness.39,40 The
`increase in suicide in HD may be due to the high
`frequency of impulsivity seen in these patients, but
`the occurrence of the suicide by our subject cannot be
`absolutely attributed to impulsivity and may possi-
`bly be related to exposure to TBZ or to other aspects
`of the underlying di