This material may be protected by Copyright law (Title 17 US. Code)
`
`The Pharmacogenomics lournal (2003) 3, 105-113
`© 2003 Nature Publishing Group All rights reserved 1470-269X/03 $25.00
`www.nature.com/tpi
`
`®
`
`ORIGINAL ARTICLE
`
`Pharmacokinetics and QT interval
`
`pharmacodynamics of oral haloperidol in poor
`and extensive metabolizers of CYPZD6
`
`2
`
`M DeSai1:2
`lE Tanussantos
`l— l—ll
`]C Corski‘
`M A,-efayene1.2
`-
`1
`Y Llu
`1 2
`Z Desta '
`DA Flockhart”
`
`‘DiY/5/°f'5 0'‘ Clinical Pharmacology, lnd/9"”
`
`Washington’ DC’ USA
`
`Correspondence:
`DA Flockhaftz Department Of Medicine/
`Division of Clinical Pharmacology, Indiana
`University School of Medicine, Wishard
`Memorial Hospital, Myers Bldg. Room
`W7123, 1001 W. 10th St., Indianapolis, IN
`46202-2879, USA.
`Tel: +1 317 630 8795
`
`Em“
`‘mm ‘
`
`3&7 ‘|::°(g_l35_ d
`0c
`'3
`lupme U
`
`Received: 26 September 2002
`Revised: 2 lanuary 2003
`Accepted: 6 january 2003
`
`.
`.
`.
`.
`ABSTRA§T
`We studied the pharmacokinetics and QT interval pharmacodynamics of a
`single 10mg dose of oral haloperidol
`in a randomized, double-blind,
`p|acebo—contro||ed, crossover trial of healthy poor (PMs) and extensive (EMS)
`metabolizers of CYPZD6. There was a statistically significant greater mean
`QTC on haloperidol (421.6i20.1 ms) than on placebo (408.4i18.5 ms,
`P—0.0053) occurring 10h post haloperidol/placebo administration. Men
`and women had similar ranges of QTC changes from placebo. Despite a
`statistically significant greater mean elimination half—life
`(19.1 i3.6 vs
`12.9i4.0 h, P:0.04) and lower mean apparent oral clearance (12.8i4.1
`vs 27.0i11.3 ml/min/kg, P:0.02) of haloperidol in CYPZD6 PMs than in
`EMS, this exposure change did not translate into marked QTC changes from
`baseline that could be considered ‘clinically important. Although the
`magnitude of the mean QTC prolongation on haloperidol relative to placebo
`isre
`factors for QT prolongation.
`The Pharmacogenomics journal
`_
`K°YW°"d5‘ hal°P°"'d°'/' CYPZD6 9°“°tYP°7 QT
`
`(2003) 3, 105 113. doi:10.1038/sj.tpj.6500160
`
`INTRODUCTION
`
`Schizophrenia is a common psychiatric disease with a lifetime prevalence of
`nearly 1% of the general population1 that is associated with an increased risk of
`premature death. A recent meta—analysis revealed that schizophrenic patients are
`1.5 times more likely of dying from all causes compared to an age— and gender-
`matched cohort of the general population? While it is not known how much of
`this excess risk can be attributed to antipsychotic-induced cardiotoxicity, it is
`clear that many antipsychotics are arrhythmogenic.3
`Among the antipsychotic drugs, haloperidol remains one of the most widely
`used worldwide. Haloperidol—induced ventricular arrhythmias of the torsades de
`pointes (TdP) type have been reported with a range of doses starting as low as
`4 mg”‘ administered over a 24 h period and as high as 825 mg5 over a 24-h period.
`Cardiac side effects at high doses likely involve excessive exposure to haloperidol.
`However, the extent to which low doses of haloperidol contribute to QT interval
`prolongation in the absence of
`risk factors6 such as age, concomitant
`medications, electrolyte imbalances,
`ischemic heart disease, or congenitally
`prolonged QT intervals is less well characterized. Prolongation of the QT interval
`is a biomarker for the malignant ventricular arrhythmia of TdP.
`In vitro cardiac electrophysiology studies that we have conducted demonstrate
`that supratherapeutic concentrations of haloperidol prolong the heart rate
`corrected QT interval (QTC) by approximately 26% in an isolated perfused feline
`heart model. The mechanism of_haloper1dol—med‘1ated_QT prolongation
`
`Vanda Exhibit 2029 - Page 1
`
`VNDA 02699851
`
`Vanda Exhibit 2029 - Page 1
`
`

`
`VV7
`
`"
`1'7
`
`1
`v
`,I‘ V
`
`VVV
`
`0.2
`
`3
`
`2cf)
`
`:5 01
`B
`an
`.0
`as
`
`A
`,3
`A
`
`
`
`subj spec
`
`fridericia
`
`bazett
`
`Figure 1 Absolute slope comparisons of each subject's (N:16)
`linear regression line between QTC vs RR using the subject-specific
`correction, Fredericia’s correction, and Bazett’s correction (based
`on placebo period data only). A lower slope suggests less potential
`for over or under correction of the heart rate corrected QT interval
`(QTc)-
`
`Subject-specific QT Correction Model
`The slopes of the QTC vs RR linear regression lines derived
`from Bazett’s, Fredericia’s, and the subject—specific heart rate
`correction formulae were compared using placebo (off-drug)
`response data. The goal of heart rate correction of the QT
`interval was to obtain a QTC vs RR linear regression line with
`a slope as close to zero as possible. As shown in Figure 1, the
`meanabsoluteslopeiSDoftheQTLinterval(ms)vsRR
`interval (ms) regression line for placebo period data using
`the subject—specific correction (0.022i0.014) was signifi-
`cantly lower than the mean absolute slopes using Freder-
`icia’s
`correction
`(0.043 450.028, P=0.04)
`or Bazett’s
`correction (0.10fi0.049, P <0.0001). Since the subject-
`specific correction generated QTC vs RR linear regression
`lines with the smallest absolute slopes, we chose to present
`study results using this correction method. Using linear
`mixed modeling for QT correction, the mean oc+standard
`error of the mean (SEM) for our study sample was 0.29 i0.02
`(range 0.23-0.38). Alpha (01) was defined as the slope of the
`log transformed QT vs RR relation using the subject—specific
`heart correction method. The terms corresponding to or in
`both the Bazett and Fredericia heart rate correction formulae
`
`were constant at 1/2 and 1/3, respectively.
`
`Pharmacokinetics
`The meaniSD pharmacokinetic parameters of haloperidol
`for the 16 subjects are shown in Table 2. As shown in Table 3,
`there was no statistically significant difference between
`females and males with respect to clearance (25.2: 12.3 vs
`23.3 : 11.8 ml/min/kg, P: 0.72), half-life
`(15.1 : 2.4 vs
`13.1 : 5.9 h, P: 0.28), and AUC (132.1 f 66.8 vs 107.1 f
`47.6 ng h/ml, P=0.50) for the 10 mg dose. Reduced halo-
`peridol, an active metabolite of haloperidol, was below
`detectable limits in the majority of subjects and therefore a
`pharmacokinetic profile of this metabolite is not reported. All
`but one of the subjects in whom this metabolite was detected
`
`Vanda Exhibit 2029 - Page 2
`
`VNDA 02699852
`
`Haloperidol phannacokinetics and pharmacodynamics
`M Desai et al
`
`®1
`
`06
`
`involves blockade of the rapidly acting delayed rectifier
`potassium channel (IKI). Haloperidol, has been shown to
`block this channel expressed in Xenopus oocytes in a
`concentration—dependent manner with an ICSO of 1 uM.8
`Despite the clear ability of haloperidol
`to bring about
`relevant changes in ion channel activity in vitro and a
`number of incriminating case reports of cardiotoxicity,9 the
`ability of therapeutic doses of haloperidol to prolong the
`QTC interval in healthy subjects without the presence of
`interacting drugs is not known.
`Cytosol reductase is the enzyme that converts haloperidol
`to reduced haloperidol, an active metabol“1te.mfl1 Reduced
`haloperidol can be oxidized back to haloperidol by cyto-
`chrome P450 isoforms CYP3A4 and CYP2D6.12*13 Multiple
`clinical studies have shown that CYPZD6 genotype influ-
`ences haloperidol and reduced haloperidol pharmacoki-
`netics.”’16 However, it is not known to what extent C YPZD6
`genotype influences haloperidol—induced QT interval phar-
`macodynamics.
`the present study was to
`The primary objective of
`mdeWr used clinical—dose
`of haloperidol to alter the QTC interval in healthy subjects in
`a prospective, randomized controlled trial. The secondary
`objectives of the study were to determine the influence of
`CYPZD6 genotype on haloperidol disposition and QT
`interval pharmacodynamics.
`
`RESULTS
`
`/\totalof16healthysubjectsparticipatedinthestudy.
`Subject demographics are presented in Table 1. The body
`mass index (BMI) ranged from 21.4 to 31.6 kg/m2 in all
`subjects with males (25.8i3.6 kg/m2) having a greater BMI
`compared to females (22.4fi 1.4 kg/mz) (P:0.036). In all,
`eight of the volunteers were CYPZD6 *1 homozygotes, two
`were *4 heterozygotes, two were *10 heterozygotes, one was
`a *17 homozygote, and three were *4 homozygotes. One
`subject who started the study dropped out because of severe
`mn
`the mean heart rate was 62.33565 beats per minute (bpm)
`and 56.3 i8.1 bpm on haloperidol and placebo, respectively
`(P:0.003 pre-Bonferonni and P:0.039 post—Bonferonni).
`The heart rate was not statistically significantly different
`between the two groups at any other time point during the
`study.
`
`Table 1 Demographics
`
`Age (years)
`Weight (kg)
`Height (cm)
`Ethnicity
`Caucasian
`African—American
`Asian
`
`Female (n=8)
`
`Males (n=8)
`
`26.9+8.0
`62.3i6.7
`166.7i7.4
`
`32.1 +4.0
`82.5 114.2
`178.5i6.2
`
`5
`1
`2
`
`4
`
`1
`
`Values are reported as the meaniSD of subjects completing the study.
`
`The Pharmacogenomics Journal
`
`Vanda Exhibit 2029 - Page 2
`
`

`
`@1
`
`07
`
`Haloperidol pharmacokinetics and pharmacodynamics
`M Desai et al
`
`were poor metabolizers (PMs). In the three PMs receiving the
`10 mg dose, the mean Cmax was 1.23 ng/ml occurring between
`24 and 48h after haloperidol administration. In the single
`extensive metabolizer (EM) in whom it was detected, the Cmax
`was 0.696 ng/ml occurring at 8 h postdose.
`The effects of CYPZD6 genotype on haloperidol pharma-
`cokinetics are shown in Figure 2. The mean terminal
`elimination half—life of haloperidol was statistically signifi-
`cantly higher
`in PMs
`(19.1 14.0 h) compared to EMS
`(12.914.0 h, P—0.04) (Figure 2a). The mean apparent oral
`clearance of haloperidol was significantly lower in PMs
`(T278141 ml/min/kg) compared to EMS (Z7.0—_—l1.3ml/
`min/kg) (P:0.02) (Figure 2b). The maximal plasma con-
`centrations of haloperidol achieved were 6.11O.3 and
`7.913.9 in PMs and EMS,
`respectively and were not
`statistically significantly different.
`
`QT Interval Pharmacodynamics
`As expected, the time averaged QTC’s1SD off drug (placebo
`only) were 4l6.81l7.9 and 408.91 16.6 ms in females and
`
`Table 2 Haloperidol pharmacokinetic parameters (n=16)
`
`Cmax (ng/mi)
`Tmax (*1)
`AUC (ng h/ml)
`Vd/F(l/kg)
`Clearance/F (ml/min/kg)
`Half-|ife(h)
`
`7.61.3.6
`29il -3
`119.6157.4
`27.0111.9
`24.3111.7
`14.114.5
`
`Values are reported as the mean1SD after administration of 10mg of oral
`haloperidol. Cmax is the maximal plasma concentration recorded in each subject.
`Tmax is the time at which Cmax occurred. AUC is the area under the plasma
`
`volume of distribution. Clearance/F is the apparent oral clearance.
`
`Table 3 Haloperidol pharmacokinetics and sex
`
`Females (n = 8) Males (n = 8)
`
`P-value
`
`AUC (ng h/ml)
`Clearance/F (ml/min/kg)
`Half-life (h)
`
`132.1166.8
`25.3112.3
`15.112.4
`
`107.1147.6
`23.3111.8
`13.116.0
`
`0.51
`0.72
`0.29
`
`Values are reported as the mean 1 SD of subjects receiving 10 mg oral haloperidol.
`AUC is the area under the plasma concentration vs time curve extrapolated to
`infinity. Clearance/F is the apparent oral clearance. P—va|ues calculated using a
`nonparametric statistical test.
`
`MSGC
`
`*4 PM
`
`EM
`27.0 a.» 11.3
`A
`‘
`
`~k*
`
`12.8 i 4.1
`I
`'
`I
`
`A
`1
`
`‘
`A A‘
`AAA
`
`‘
`
`60
`
`time (hr.)
`
`80
`
`100
`
`b 9 s
`T
`8 ’ ll
`7 -1
`6 '
`-
`E 5 1
`E’ 4 —
`3 ‘
`4
`2 —
`
`J,
`
`1
`U
`
`‘
`
`0
`
`*
`
`N
`1
`.\
`l“ x. \
`N‘.
`V
`
`1
`20
`
`W 30
`l" 25
`" 20
`-— 15
`0'
`10 8
`E
`5
`
`'
`
`/
`
`
`
`~
`
`‘~T— _ _.
`.
`.
`so
`40
`time post haloperidol (hr)
`
`1
`so
`
`N
`
`0
`1, _5
`-TU
`100
`
`Q3
`O.C
`
`b
`
`C”
`§
`'—
`E
`75'
`
`50
`40
`
`20
`
`10
`
`0
`
`*4 PM's
`
`EM's
`
`(a) Mean QT(1SD (ms) on haloperidol and placebo as a
`Figure 3
`function of time post haloperidol/placebo administration. (b) Mean
`QTC change1SD (ms) and mean plasma haloperidol concentra-
`(a) Distribution of half—lives (h) as a function of CYPZD6
`Figure 2
`tions1SD (ng/ml) as a function of time posthaloperidol adminis
`tration. The QTC change is defined as QTC on haloperidol minus the
`genotype. (b) Distribution of apparent oral clearances (ml/min/kg)
`QT‘ on placebo at the corresponding time point. *P:0.00S3 (after
`as a function of CYPZD6 genotype (mean half—life and clearance/
`0:92. .
`T& :*P 0.—94;**P
`
`www.nature.com/tpj
`
`Vanda Exhibit 2029 - Page 3
`
`VNDA 02699853
`
`Vanda Exhibit 2029 - Page 3
`
`

`
`3 40
`
`30
`
`20
`
`TO
`
`8
`E
`
`-I
`-I
`
`I
`I.
`I
`
`:
`I
`I
`I
`I
`I
`
`A
`
`A
`A
`A
`
`A:
`AA::
`‘
`A
`A
`
`0
`
`I H
`
`aloperidol
`
`Placebo
`
`b
`
`Haloperidol phamiacokinetics and pharmacodynamics
`M Desai et al
`
`®1
`
`08
`
`males, respectively (P:0.001), which is consistent with
`findings in the medical literature that suggest females in
`general have longer QT intervals than males in the absence
`of drug therapy.
`Figure 3a shows the mean QTC’s:SD in the treatment and
`placebo groups at each time point during the study. At the
`10 h time point,
`the mean QTc’s were 421.6:20.1 and
`408.41 18.5 ms on haloperidol and placebo, respectively
`(P=0.00041 pre-Bonferonni, P:0.0053 post-Bonferonni).
`At the 4- and 6-h time points, there were trends towards a
`greater mean QTC on treatment
`than placebo but after
`Bonferonni’s correction the trend was nullified. Figure 5b
`shows both the haloperidol—induced mean QTC changes
`from placebo and the mean haloperidol plasma concentra-
`tions as a function of time post dosing. The QTC change is
`defined as the QTC on treatment at a given time point less
`the QTC on placebo at the corresponding time point. Males
`and females had significant overlap in their maximal QT:
`changes from placebo as shown in Figure 4a. There was also
`significant overlap in QTC changes from placebo at Tmax (the
` -
`tions were achieved) (Figure 4b). Similarly among PMs and
`
`3 60
`
`so
`
`40
`
`U)
`55
`E 30
`
`20-
`
`'
`
`I
`I.
`
`,:I
`
`A
`:
`1
`‘A
`
`Sex
`
`b so
`
`§
`E
`
`-30
`
`Made
`
`Femaue
`
`Male
`
`Female
`
`§
`5
`
`A
`
`
`
`C
`
`8
`E
`
`Genotype
`d 30
`
`20
`
`10
`
`§
`E O
`
`.10
`
`‘20
`_30
`
`0
`
`PM's
`
`EM's
`
`PM's
`
`EM's
`
`Figure 4 Haloperidol-induced QTC changes from placebo as a
`function of sex and genotype.
`(a) and (c) show maximal QTC
`changes in milliseconds from placebo occurring at any time
`postdose. (b) and (d) QT: changes in milliseconds from placebo
`occurring at Tmax (time point where maximal haloperidol concen-
`
`
`The Pharmacogenomics Journal
`
`Haloperidol
`Placebo
`Figure 5 QTC changes from baseline: haloperidol vs placebo. (a)
`Maximal QT: change in milliseconds from baseline occurring at any
`time postdose.
`(b) QTC change in milliseconds from baseline
`occurring at Tmax (time point where maximal haloperidol concen-
`trations were achieved).
`
`EMS of CYPZD6, there was significant overlap between the
`QTC changes from placebo as shown in Figure 4c and 4d.
`There was significant overlap in the maximal QTC change
`from baseline (time 0 time point) in the treatment and
`placebo groups as shown in Figure 5a. Similarly, significant
`overlap in the QTC change from baseline occurred at the
`time point when maximal haloperidol plasma concentra-
`tions were achieved as shown in Figure 5b.
`The maximal change in QTC relative to baseline observed
`in a single individual at any time post haloperidol admin-
`istration was 8.8"/0. Similarly, the maximal change in QTC
`relative to baseline observed in any sub}ect
`receiving
`
`Vanda Exhibit 2029 - Page 4
`
`VNDA 02699854
`
`Vanda Exhibit 2029 - Page 4
`
`

`
`@1
`
`09
`
`Haloperidol pharmacokinetics and pharmacodynamics
`M Desai et al
`
`occurring 10h post oral haloperidol administration. This
`difference could not solely be the result of intraobserver
`variability in QT interval assessment because there was only
`a mean 1.3 ms difference at time 0 (baseline) in both the
`placebo and haloperidol
`treatment periods. This 1.3—ms
`difference likely reflected the sum of both intra-observer
`variability and intraindividual variability in the QT/QTC and
`accounted for about 10% of the change seen at the 10-h
`time point.
`Our results differ from other studies in the literature that
`
`suggest haloperidol does not cause statistically significant
`Qlc prolongation when usedfiin relatively low doses. l<ulop et
`ails reported a nonsignificant QTC prolongation of <4 ms at
`the end of 6 weeks of treatment with oral haloperidol (doses
`up to 10 mg/day) in patients with Tourette’s SyI1dIOII1€.l9
`The mean dose used in Fulop’s study was approximately
`5 mg a day. There are major differences in the two study
`designs. The obvious difference was that our study was a
`single—dose study whereas Fulop’s was a multidose study. In
`our study, all subjects receiving haloperidol also served as
`-
`X
`-
`-
`number of subjects had both a placebo and treatment
`period. Additionally, we had a comprehensive placebo
`period during which ECG sampling was intensively per-
`formed,
`therefore allowing us to monitor
`the natural
`fluctuations of QTC in the absence of treatment. Intense
`placebo period ECG sampling was not performed in Fulop’s
`study. Also in our study, we acquired ECGs at multiple
`prespecified time points post haloperidol administration
`allowing—ustodeteetQTLehangesthatweredelayedfroi:n
`peak plasma concentrations. In Fulop’s study, acquiring only
`a single ECG at the end of the study may have caused a peak
`QT effect
`that was significantly delayed from the peak
`plasma concentration to be missed. Additionally using a
`range of doses up to 10 mg may have diluted the power to
`detect an effect at any specific dose (eg 10 mg). Another
`possibility of a lack of effect in Fulop’s chronic dose study
`could be tolerance to IK, blockade. It
`is unclear if this
`phenomenon occurs with haloperidol or other drugs. In
`general, we believe our study had greater sensitivity to
`detect an effect of haloperidol on the QTC interval than did
`Fulop’s study.
`We found a small but statistically significant effect of
`C YPZD6 genotype on haloperidol pharmacokinetics in that
`the terminal elimination half—life was greater and the
`apparent oral clearance was lower in PMs than in EMs.
`These findings were consistent with findings from other
`studies.”*‘4 However, the exposure differences attributed to
`C YPZD6 genotype were not sufficient to produce substantial
`haloperidol—induced QTC pharmacodynamic changes in PMs
`relative to EMs. A likely reason for this observation could be
`that CYPZD6 is not exclusively responsible for haloperidol
`disposition. It is known that several P4505 and non-P450
`enzymes are involved in this process, thus making it difficult
`for a deficiency in any particular metabolic pathway to
`markedly influence QT interval pharmacodynamics. Cyto-
`solic ketoreductase is the enzyme responsible for conversion
`of haloperidol to reduced h'cll0p€I1dOl.I0’_H In vltro studies
`
`for significant
`illustrating potential
`placebo was 7.2%,
`variability of this measurement
`in the absence of any
`treatment.
`
`There was poor correlation between plasma haloperidol
`concentrations and QTC change from placebo during the
`first 10h of the study (R2:O.OO7, P>O.30)
`(figure not
`shown).
`
`Adverse Effects
`
`No subject was discontinued from the study because of
`
`
`
`because of severe anxiety and restlessness starting 4 h after
`receiving haloperidol. The most common side effects seen
`with haloperidol were anxiety and restlessness of variable
`intensity that occurred in 12 of 16 subjects (75%) complet-
`ing the study. Other less common side effects that occurred
`at a frequency of between 10 and 40% were difficulty
`concentrating, feeling tired or sleepy, decreased appetite,
`dry mouth, blurred vision, dystonia, and vivid dreams.
`
`.
`experienced
`’I—m;ee
`after dosing and were successfully treated with diphenylhy-
`dramine 25 mg orally.
`Subjects experiencing dystonia requiring diphenylhydra—
`mine did not differ significantly in haloperidol pharmaco-
`kinetic parameters compared with those not experiencing
`these side effects. Subjects experiencing dystonia showed a
`mean clearance of 15.3i4.1 ml/min/kg, while those not
`experiencing this side effect showed a mean clearance of
`26.02%-12.4 ml/min/kg ('P:0.071). The mean Cm, of
`haloperidol
`in
`subjects
`experiencing
`dystonia was
`9.4i4.2 ng/ml, while the Cmax in subjects not experiencing
`dystonia was 7.1 fi3.4 ng/ml (P=O.35). Similarly, there was
`no statistically significant difference in the mean plasma
`concentration vs time area under the curve between the two
`
`groups (158€g94 vs 111 i417 rig h/ml, P=O.35). The CYPZD6
`genotypes for the three subjects who experienced dystonia
`were *4 heterozygote, *l0 heterozygote, and *4 homozygote.
`
`DISCUSSION
`
`We conducted a study of the effects of routinely used, low
`doses of oral haloperidol on the electrocardiograpliic QT
`interval pharmacodynamics in healthy volunteers not on
`concomitant medications. To improve the mechanistic
`understanding of our data, we determined the pharmacoki-
`netics of haloperidol
`in each subject and the effect of
`CYPZD6 genotype on pharmacokinetics and QT interval
`pharmacodynamics. We chose to study a healthy population
`because of the potential for multiple, confounding, drug
`and disease interactions in a patient population and because
`of the specific potential for ion channel variants to occur in
`the hearts of patients with schizophrenia. Potassium
`channel variants have been reported in the brains of
`schizophrenics.” Single doses rather than multiple doses
`were used because of the intolerability of the latter study
`design in normal, healthy volunteers.
`Our data showed a statistically significant mean QTC
`prolongafion of approximately f3ms relative to placebo
`
`www.nature.com/tpj
`
`Vanda Exhibit 2029 - Page 5
`
`VNDA 02699855
`
`Vanda Exhibit 2029 - Page 5
`
`

`
`known. Given the large medical, social, and economic
`burden of this disease, the benefits of haloperidol signifi-
`cantly outweigh the relatively small risk of drug—induced
`TDP when used in low doses in patients without risk factors
`for QT prolongation.
`
`METHODS
`
`Haloperidol phamiacokinetics and pharmacodynamics
`M Desai et al
`
`®1
`
`10
`
`have shown that CYP3A can oxidize reduced haloperidol
`back to haloperidol and can N—dealkylate haloperidol to the
`inactive metabolite 4—(p—chlorophenyl)—4—hydroxypiperidine
`(CPHP) or aromatize haloperidol to the inactive metabolite
`4—(4-chlorophenyl)—1—[4—(fluorophenyl)-4-oxybutyl]-pyridi-
`nium (HPP").2°'21 Yasui et al” showed that schizophrenic
`patients prescribed haloperidol 12-14 mg/day who were
`coadministered the CYP3A4 inhibitor,
`itraconazole, had
`significantly higher concentrations of the parent and active
`metabolite with an associated increase in the incidence of
`
`Subjects
`We studied 16 healthy, nonsmoking subjects using single
`10 mg doses of oral haloperidol. All subjects were aged
`between 21 and 41 years. Subjects gave written informed
`consent prior to study participation. The study protocol was
`approved by the Institutional Review Board of Georgetown
`University Medical Center.
`Subjects were included if there were no clinically sig-
`nificant abnormalities in their medical histories, physical
`examinations, hematologic tests, blood chemistries (includ-
`ing K’“, Mg2" Ca2+), or urine toxicologic tests. Female
`subjects were required to have a negative serum [3—HCG 2
`days prior to study initiation. Subjects were excluded if they
`had a screening electrocardiogram (ECG) with a baseline
`QTC greater than 440 ms by Bazett’s correction or a history of
`any cardiac disease, including myocardial infraction, valv-
`ular heart disease, or cardiomyopathy. Subjects were also
`excluded if they were actively using any prescription or
`nonprescription drugs including oral contraceptives.
`All study subjects were required to refrain from alcohol,
`grapefruit juice, and caffeine—containing beverages begin-
`ning 48h prior to the study and throughout the study
`duratron.
`
`neurological side effects. A role for CYP1A2 has been
`suggested in haloperidol clearance but is controversial.2324
`The accumulated evidence suggests that multiple enzymes
`are involved in haloperidol metabolism, therefore diluting
`any specific effect of CYPZD6 genotype on QTC pliarniaco—
`dynamics in our study.
`formed from haloperidol by
`is
`Reduced haloperidol
`cytosolic ketoreductase and represents one of the major
`routes of haloperidol metabolism.
`In addition,
`there is
`evidence that reduced haloperidol is partially converted to
` mmm—
`idol contributes significantly to QT interval pharmacody—
`namics. Reduced haloperidol is a less potent inhibitor of the
`1K, channel than its parent with an IC50=2.6 uM.8 Also, it
`has been shown that when haloperidol
`is administered
`orally, plasma concentrations of the parent compound are at
`least two to three times higher than the metabolite.” Both
`the higher IC50 and lower plasma concentrations achieved
`argue against a contribution of reduced haloperidol
`in
` ' the Q31} 1 . Home’V61‘, .‘
`studie‘ 5
`involving administration of reduced haloperidol would be
`necessary to confirm this hypothesis.
`Study Design
`This
`study was conducted in Georgetown University's
`Overall, our data clarify a number of important issues
`General Clinical Research Center (GCRC). This protocol was
`related to the arrhythmogenicity of haloperidol. First, the
`a randomized, double-blind, placebo—controlled, single-dose
`magnitude of the mean QTC change brought about by
`crossover clinical trial. There were two study periods with
`commonly used low doses of haloperidol
`is statistically
`each period lasting 4 days. During the first 24 h of each study
`significant but clinically small. While this magnitude of
`period, subjects were required to stay in the GCRC overnight.
`change will not manifest clinically in a majority of patients,
`there are individuals in whom the magnitudes of changes There was a 3-day washout between the two study periods.
`we detected may become important when other risk factors
`Subjects were randomly assigned to receive either placebo or
`for QT prolongation or TdP are present
`(eg bradycardia,
`haloperidol (Geneva Pharmaceuticals Lot # 114748, Broom-
`electrolyte abnormalities, genetic predisposition, drug inter-
`field, CO, USA) at 8 am during the first study period and the
`actions, etc.). Second, although the participation of C YPZD6
`alternative treatment at the identical time during the second
`genotype in haloperidol disposition was confirmed,
`the
`study period. Subjects were observed to randomly receive
`pharmacokinetic changes observed were not sufficient to
`placebo or haloperidol on an empty stomach after an
`bring about clinically important pharmacodynamic con-
`overnight fast and were given breakfast approximately 1 h
`sequences. Based on our findings, we believe that
`the
`after dosing. Meals during the GCRC stay were standard with
`routine acquisition of a screening electrocardiogram is
`no citrus fruits or caffeine—containing beverages.
`warranted prior to use of low doses of oral haloperidol in
`After a 24-h stay in the GCRC, subjects were discharged
`otherwise healthy subjects not on concomitant medica-
`and returned every 12 h for the remainder of each study
`tions. An electrocardiogram provides a relatively cheap,
`period. Blood samples for analysis of haloperidol concentra-
`tions were collected before and 2, 4, 6, 8, 10, 24, 36, 48, 60,
`noninvasive, time—efficient, and simple to use screening tool
`that can be performed in an inpatient or outpatient setting
`72, 84, and 96 h after administration of either placebo or
`haloperidol. ECGs were performed immediately prior to
`that may help identify patients with risk factors such as
`genetic potassium channel variants and decreased repolar—
`each blood draw, for a total of 13 recordings in the placebo
`period and 13 recordings in the treatment period. ECGs were
`ization reserve. Currently, there is no standard method for
`screening such individuals and the frequency with which
`obtained with the subject resting supine for approximately
`these genetic variants occur in the general population is not
`T5 min prior to acquisition using a MA? 3 SUUU EC? TG machine
`
`The Pharmacogenomics Journal
`
`Vanda Exhibit 2029 - Page 6
`
`VNDA 02699856
`
`Vanda Exhibit 2029 - Page 6
`
`

`
`@1
`
`11
`
`Haloperidol pharmacokinetics and pharmacodynamics
`M Desai et al
`
`Table 5 Thermocycling conditions for *3, *4, *6, *8, *10, *17
`
`Conditions for
`
`Conditions for
`
`No. of
`
`*3, *4, *6, *8 alleles
`
`*10, *17 alleles
`
`cycles
`
`Initial denaturation
`
`94°C >< 2 min
`
`Amplification
`
`Terminal extension
`
`94°C >< 30$
`60°C><10s
`72°C >< 1 min
`72°C >< 7min
`
`94°C >< 2 min
`
`94°C >< 30$
`58°C><10s
`72°C X 1 min
`72°C >< 7min
`
`1
`
`30
`
`1
`
`as two separate mixtures of 25 pl each. The first of these was
`a mixture containing 17.5 pl of water, 1.7 pl of dNTP’s, 0.3 pl
`each of forward and reverse primers and 5 pl of genomic
`DNA (~50O ng). The second 25 pl mixture contained
`19.25 pl of water, 5 pl of buffer (NE Biolabs #3), and DNA
`Taq Polymerase 0.75 pl. Both mixtures were prepared on ice
`and mixed by gentle Vortexing just prior to thermocycling.
`The details of the thermocycling conditions of the LT-PCR
`productare—slToWninTable4.TheLT-PCRprodtrctWas
`diluted with 200 pl of Tris—EDTA buffer and stored at —20°C
`for later use in nested/allele specific PCRs.
`The nested or allele specific PCRs for *3, *4, *6, *8, *10, *17
`were all prepared at a volume of approximately 25 pl. Each
`mixture contained 22.5 pl of Platinum Supermix, 1.5 pl of
`distilled water, 1 pl of LT-PCR product, and 0.25 pl each of
`forward and reverse primers. The details of the thermo-
`cycling conditions for alleles *3, *4, *6, *8, *10, *17 are
`showninTable5.—AHf9rward—andrevepseprirne1LsrfortheLT-
`PCR product and allele—specific PCR products as well as allele
`specific restriction enzymes are listed in Table 6.
`
`Pharmacokinetics
`
`Approximately 7 ml of whole blood was collected into a red
`top Vacutainer*"7‘
`tubes and immediately separated by
`centrifugation at 4000 >< g for 10 min followed by transfer
`into a cryogenic tube and refrigeration at —20°C. Plasma
`concentrations of haloperidol and reduced haloperidol were
`
`(GE Medical Systems, Milwaukee, WI, USA). We acquired 12
`lead ECGs at a paper speed of 50 mm/s and amplitude of
`20 mm/mV. The electrocardiograms were printed in a
`format
`that displayed rhythm strips from all 12 leads
`simultaneously in order to facilitate identification of the
`earliest Q wave and latest T wave.
`
`Analysis of Blood Samples
`Genotype determination
`
`Blood samples for genotype determination were collected in
`§[aC”ta]'DeI:i<,
`tubes (BeCtOD_D1CkjnsOD Erankhn lakes NJ7
`USA) containing sodium heparin. The blood was then
`transferred into a cryogenic vial (Corning, Cambridge, MA,
`USA) for storage at —20°C. Genomic DNA was extracted
`from the leukocyte portion of whole blood using a QIAamp
`DNA blood Midi Kit ‘"'(Qiagen, Valencia, CA, USA). Screen-
`ing for (,'YP2l)6 alleles *1, *3, *4, *6, *8, *10, and *17 was
`performed as previously described“ using amplification of a
`4.7 kilo base pair (kbp) fragment containing all nine exons,
`followed by nested polymerase chain reactions (PCR) for
`each genetic variant tested.
`the 4.7 kbp long
`The final amplification mixture for
`template PCR product (LT-PCR) was 50 pl. It was prepared
`
`Table 4 Thermocycling conditions for long template PCR
`product
`
`Initial denaturation
`
`Amplification
`
`Amplification
`
`Conditions for
`
`ET-PCR product
`
`94°C x 2 min
`94°C >< 10s
`65°C >< 30s
`68°C >< 4 min
`94°C X 10 s
`65°C x 305
`
`Terminal extension
`
`68°C >< 4min (adding 20s
`to each subsequent cycle)
`68'C >< 7mm
`
`No. of
`
`cycles
`
`1
`10
`
`25
`
`1
`
`Table 6 Allele specific primers and restriction enzymes
`
`Primers (forward(F) and reverse(R))
`
`Restriction enzyme
`
`LT-PCR product
`
`*3
`
`*4, *6, *8
`
`*1 O
`
`*17
`
`5’-GGCCTACCCTGGGTAAGGGCCTGGAGCAGGA-3’ (F)
`5’— ’ (R)
`5’-GCTGGGGCCTGAGACTT-3’ (F)
`5’-GGCTGGGTCCCAGGTCATAG3’ (R)
`5’-CCTGGGCAAGAAGTCGCTGGACCAG-3’ (F)
`5’-GAGACTCCTCCGTCTCTCG-3’ (R)
`5’-TCAACACAGCAGGTFCA-3’ (F)
`5’-CTGTGGTTTCACCCACC-3’ (R)
`5’-CTCGTGCTCAATGGGCTGGCGGCCGTGCGCGAGGCG-3’ (F)
`5’-CTCACCCCTCTCGCCCGG'lTCTTl'GG-3’ (R)
`
`BscIA1
`
`*4, *6 BstN1
`*8 ll/lspl
`HpH1
`
`Fok1
`
`This table provides a list of both the fon/vard and reverse primers used in genotyping for various alleles of CYP2D6 along with the associated allele—specific restriction
`enzymes.
`
`www.nature.com/tpj
`
`Vanda Exhibit 2029 - Page 7
`
`VNDA 02699857
`
`Vanda Exhibit 2029 - Page 7
`
`

`
`Haloperidol phannacokinetics and pharmacodynamics
`M Desai et al
`
`®1
`
`12
`
`analyzed using liquid chromatography/mass spectrometry
`(LC/MS)
`(Finnigan AQA, San Jose, CA, USA). A 0.5 ml
`sample of plasma was mixed with 20 pl of internal standard
`[1 pg/ml of chlorinated haloperidol (Sigma RBI, St Louis,
`MO, USA)] and 250 pl of
`1 M NaOH/glycine buffer,
`pH : 11.3, in a 13 >< 100 mm disposable culture tube (Fischer
`Scientific, Pittsburgh, PA, USA). To the mixture, 4ml of
`cyclohexane/methylene chloride (7 : 3, V : V) was added. All
`samples were mixed vigorously for 15 s using a platform
`mixer. The samples were then centrifuged in a Sorvall RT
`6000D3"“ centrifuge (Sorvall, Wilmington, DE, USA) at
`2800 rpm for 5 min. The organic phase was transferred to
`a clean disposable culture tube and evaporated to dryness
`using a Savant SpeedVac Concentrator“ (Thermo—Savant,
`Faririingdale, NY, USA). The residue was reconstituted with
`120 pl of mobile phase and the contents were transferred
`into a polypropylene tube inside an autosampler vial. A
`volume of 40 pl was injected into the LC—MS. A Luna 3 p CN
`100 x 2.00 mm column (Phenomenex, Torrance, CA, USA)
`was used. Two mobile phases were used: 10 mM ammonium
` 20”0) arrd 101rrM
`am