Influence of CYP2D6 polymorphism on the
`pharmacokinetics and pharmacodynamics
`of tolterodine
`
`is involved in the metabolism of toltero-
`OCjectiPe: To determine whether cytochrome P450 2D6 (CYP2D6)
`dine by investigating potential differences in pharmacokinetics and pharmacodynamics
`(heart rate, accom-
`modation, and salivation) of tolterodine and its 5-hydroxymethyl metabolite between poor metabolizers
`and extensive metabolizers of debrisoquin
`(INN, debrisoquine).
`received 4 mg
`Methods: Sixteen male subjects (eight extensive metabolizers and eight poor metabolizers)
`tolterodine by mouth
`twice a day for 8 days followed by a single intravenous
`infusion of 1.8 mg toltero-
`dine for 30 minutes after a washout period. Doses were given as the tartrate salt. The pharmacokinetics of
`tolterodine and 5-hydroxymethyl metabolite were determined, and the pharmacodynamics were measured.
`Results: The mean systemic clearance of tolterodine was significantly
`lower (p < 0.001) among poor metab-
`olizers (9.0 + 2.1 L/hr)
`compared with extensive metabolizers
`(44 * 13 L/hr),
`resulting
`in a fourfold
`longer elimination half-life
`(p < 0.001). The terminal half-life of the 5-hydroxymethyl metabolite
`(2.9 =
`0.4 hours) was slightly longer than that of the parent compound
`(2.3 * 0.6 hours) among extensive metab-
`olizers, but the 5-hydroxymethyl metabolite was undetectable
`in the serum of poor metabolizers. Only
`minor differences in pharmacodynamic effects after tolterodine dosage were observed between the groups.
`Tolterodine
`caused a similar decrease in salivation
`in both panels. The decrease occurred when the con-
`centration of unbound
`tolterodine and 5-hydroxymethyl metabolite among extensive metabolizers was
`comparable with that of tolterodine among poor metabolizers.
`Conclusions: Tolterodine
`is extensively metabolized by CYP2D6 with high specificity. Despite the effect
`on pharmacokinetics,
`the CYl?2D6 polymorphism
`does not appear to be of great importance
`in the
`antimuscarinic effect, probably because of the additive action of parent drug and active metabolite.
`(Clin
`Pharmacol Ther 1998;63:529-39.)
`
`Al&n, MD,
`BSc, Per Dal&, MD, Gunnar
`Niclas Brynne,
`Johan Gabrielsson,
`PhD Uppsula and Huddinge Sweden
`
`Leif Bert&son,
`
`PhD, and
`
`Large interindividual variation in drug response and
`therapeutic outcome
`is not uncommon, and a leading
`contributing
`factor is a corresponding variation in the
`
`at
`
`and
`Pharmacia
`Pharmacology,
`of Clinical
`the Department
`From
`and
`the Department
`of Medical
`Laboratory
`Upjohn
`AB, Uppsala,
`Sciences
`and Technology,
`Division
`of Clinical
`Pharmacology
`the Karolinska
`Institute,
`Huddinge
`University
`Hospital,
`Huddinge.
`Supported
`by Pharmacia
`& Upjohn
`AB, Sweden.
`of
`Presented
`in part at the Sixth European Meeting
`Society
`for
`the Study
`of Xenobiotics,
`Gothenburg,
`30-July
`3, 1997.
`Oct. 7, 1997; accepted
`Received
`for publication
`Brynne,
`Department
`Reprint
`requests:
`Niclas
`ogy, Pharmacia
`& Upjohn
`AB,
`SE-751
`E-mail:
`niclas.brynne@eu.pnu.com
`Copyright
`0 1998 by Mosby,
`Inc.
`0009-9236/98/$5.00 + 0 13/l/87981
`
`the
`
`International
`Sweden,
`June
`
`Dec. 4, 1997.
`of Clinical
`Pharmacol-
`82 Uppsala,
`Sweden.
`
`is
`drugs. One example
`to metabolize
`capability
`(CYP2D6)
`polymorphism
`cytochrome P450 2D6
`(debrisoquin
`[INN, debrisoquinel-sparteine
`hydroxy-
`lase), which exhibits an incidence of poor metabolism
`of about 7% among white persons.1 A large number of
`drugs, such as antidepressants, neuroleptic agents,
`antiarrhythmic
`agents, several P-adrenergic
`receptor
`antagonists and some opioids, have been shown to be
`metabolized by the CYF’2D6 enzyme.233 Individualiza-
`tion of dosage is necessary for some of these drugs.4
`Moreover,
`if a drug is metabolized by CYP2D6
`to an
`active metabolite, the activity of the enzyme may be an
`important determinant of the effectiveness of treatment.
`Certain drugs, such as codeine5 and encainide,h are
`activated by CYP2D6. Therefore poor metabolizers
`may be nonresponders
`to such therapy. Conversely,
`with drugs
`that are inactivated by CYP2D6, poor
`
`529
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 1
`
`

`
`530
`
`Brynne et al.
`
`CLINICAL
`
`PHARMACOLOGY
`
`& THERAPEUTICS
`MAY 1998
`
`Tolterodine
`
`N-dealkylated tolterodine
`[5-HM]
`tolterodine
`5-hydroxymethyl
`Fig. 1. Tolterodine, (R)-N,N-diisopropyl-3-(2-hydroxy-5-methylphenyl)-3-phenylpropanamine and
`the primary metabolites of tolterodine.
`
`metabolizers may accumulate the drug and experience
`adverse effects at “normal” dosages.
`Tolterodine is a new antimuscarinic drug for the man-
`agement of overactive bladder with symptoms of fre-
`quency or urge incontinence.7 Preclinical studies have
`shown that tolterodine has high antimuscarinic potency
`in guinea pig and human detrusor muscle and displays
`favorable selectivity for the urinary bladder over sali-
`vary glands
`in vivo.8 Studies have suggested
`that
`tolterodine exerts a marked inhibitory effect on blad-
`der function
`among both healthy volunteers9 and
`patients with urinary urge
`incontinence.lO Among
`healthy volunteers,
`tolterodine was rapidly absorbed,
`with a high first-pass metabolism. This led to consid-
`erable interindividual
`variation
`in serum concentra-
`tion.” Two hepatic oxidative metabolic pathways for
`tolterodine have been identified-hydroxylation
`and
`N-dealkylation
`(Fig. 1)” The concentration-effect rela-
`tion suggests a pharmacologically active metabolite. Pre-
`clinical studies have shown that the 5-hydroxymethyl
`metabolite (5HM) of tolterodine (PNU-200577) is phar-
`macologically active and equipotent compared with
`tolterodine
`in vitro.12 Stahl et al.9 reported results for a
`healthy volunteer with a half-life of tolterodine of 15
`hours, which is six times longer than average. This per-
`son subsequently was phenotyped with debrisoquin and
`found to be a poor metabolizer (data on file). This find-
`ing, along with a number of unpublished observations,
`
`indicates that CYP2D6 may be involved in the metab-
`olism of tolterodine. The aim of this study was to inves-
`tigate potential differences
`in pharmacokinetics and
`pharmacodynamics
`(effect on heart rate, accommoda-
`tion, and salivation) of
`tolterodine
`and
`its 5-HM
`between poor metabolizers and extensive metabolizers
`of debrisoquin.
`
`METHODS
`Subjects. Sixteen healthy male volunteers were
`recruited
`for the study from more than 1000 healthy
`Swedish
`subjects
`previously
`phenotyped
`with
`debrisoquin.13 Eight volunteers were classified as
`poor metabolizers of debrisoquin
`(metabolic
`ratio
`>12.6) and eight were extensive metabolizers
`(meta-
`bolic ratio ~1.0). The lower metabolic
`ratio of less
`than 1.0 was chosen arbitrarily
`to exclude intermedi-
`ate metabolizers. The mean (*SD) demographic char-
`acteristics for the extensive metabolizers were as fol-
`lows: age, 29 + 7 years; body weight, 76 f 7 kg; and
`height, 1.79 f 0.06 m. The mean characteristics for the
`poor metabolizers were as follows: age, 30 + 8 years;
`body weight, 76 f 8 kg; and height, 1.81 + 0.04 m. The
`two panels were matched for age (-c2 years) and body
`weight (220%). The study was approved by the ethics
`committee of Huddinge University Hospital, and each
`volunteer gave witnessed verbal
`informed
`consent
`before the study.
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 2
`
`

`
`CLINICAL
`VOLUME
`
`P HARMACOLOGY
`63, NUMBER 5
`
`& THERAPEUTICS
`
`Brynne et al.
`
`531
`
`Study design. Volunteers received 4 mg oral tolterodine
`twice a day for 8 days, followed by a single intravenous
`infusion of 1.8 mg tolterodine
`for 30 minutes after a
`washout period (at least 7 days). Doses of tolterodine are
`given as the tartrate salt. Each volunteer fasted overnight
`before the tirst (day 1) and last (day 8) days of oral admin-
`istration of tolterodine and before the intravenous infu-
`sion. Smoking and consumption of alcohol and caffeine-
`containing beverages were prohibited the day before dos-
`ing and for 24 hours afterward.
`Assessment. On days 1 and 8, venous blood samples
`were taken before administration of tolterodine; at 10,20,
`30, and 45 minutes; and at 1, l%, 2,4,6, 8, 10, 12, and 24
`hours (24 hours on day 8 only) after administration.
`Among poor metabolizers, the 6- and lo-hour samples
`were changed to 32 and 48 hours on day 8. Blood sam-
`ples were obtained before the start of the intravenous infu-
`sion; at 5, 10,20, and 30 minutes during the infusion; and
`at 5, 10,20,30 minutes and 1,2,4,6, 8, 10, and 12 hours
`after termination of the infusion. The 6- and lo-hour sam-
`ples were replaced by samples at 24 and 26 hours among
`poor metabolizers. Urine was collected during the last
`interval of oral administration (0 to 12 hours) and after the
`intravenous infusion.
`One day before drug administration, baseline record-
`ings of pharmacodynamic measurements
`(heart rate,
`supine blood pressure, visual accommodation, and sali-
`vation) were performed
`for 8 hours. Heart rate and
`blood pressure were measured twice before tolterodine
`administration and repeated in connection with blood
`sampling. Recordings were performed simultaneously
`with an automatic, noninvasive, digital blood pressure
`meter (UA-751; A & D Company Ltd., Tokyo, Japan).
`An electrocardiogram was obtained, and near point of
`vision and salivation were measured
`twice before
`tolterodine administration
`on days 1 and 8 and at 10
`and 30 minutes and 1, 2, 4, 8, and 12 hours after drug
`administration. Recordings also were obtained before
`infusion; at 30 minutes during infusion; and at %, 4, 8,
`and 12 hours after cessation of the infusion.
`Electrophysiologic measurements were obtained with
`a computerized 12-lead electrocardiograph
`(Megacart;
`Siemens Elema, Solna, Sweden), and QT, QT,, PQ, and
`QRS duration were calculated automatically. Near point
`of vision was determined according to the Royal Air Force
`neat-point rule by means of a convergence meter (Clement
`Clarke, International Ltd., Harlow, England). Basal whole-
`mouth salivation was measured by means of absorption
`into three preweighed cotton rollsI
`(Celluron; Hartmann,
`Heidenheim-Brentz, Germany).
`Adverse events were assessed by means of sponta-
`neous reports,observation,
`and questioning at regular
`
`intervals. The intensity of the adverse event was rated
`on a three-point scale (mild, moderate, or severe) by a
`research clinician. Laboratory values were assessed
`before and at the end of the study, and vital signs data
`were screened for trends.
`Analytical method. Determination
`of tolterodine
`and 5-HM
`levels in blood, serum, and urine was per-
`formed with a specific and sensitive capillary gas
`chromatography-mass
`spectrometry assay.15 Extrac-
`tion of
`the analytes was performed with
`liquid-
`liquid or solid-phase extraction before derivation
`with a silyl-reagent. The derivatives were quantified
`by means of selected ion-monitoring mass spectrom-
`etry with deuterium-labeled
`internal standards and a
`single-level calibration
`curve. With
`this technique
`the accuracy (interday and intraday)
`for both ana-
`lytes was within 87% to 110% over the range from
`0.90 to 210 nmol/L. Precision was better
`than 9%.
`Urinary concentrations of tolterodine and 5-HM also
`were assessed after
`incubation
`of samples with
`P-glucuronidase
`(Boehringer Mannheim,
`GmbH,
`Mannheim, Germany).
`Data analysis. All data are expressed as mean + SD,
`except when
`indicated otherwise.
`In the regression
`analysis of tolterodine data (performed with PCNON-
`LIN
`[version 4.1]‘6), equations 1 and 217 were fitted
`simultaneously to intravenous and oral data, as follows:
`
`c, = Rinf
`‘C
`
`i
`i=l
`
`[ -~.[l-e~;?.e-i;tf]
`1
`
`(1)
`
`k, . F . Dose
`c,, = ~. ~
`~
`
`vc
`
`ci ~
`kahi
`
`. e-h,’
`
`t)
`
`(2)
`
`I
`
`in which Rinf is the infusion rate, V, is the apparent vol-
`ume of the central compartment, n is the number of
`exponential terms, Ci is the fractional intercept, & is the
`corresponding rate constant, t is time, and T is time dur-
`ing the infusion and then becomes a constant, tinf (dura-
`tion of infusion), after the cessation of the infusion, k,
`is the absorption rate constant, and F is the extent of
`bioavailability. A weighting scheme of l/8,
`in which a
`constant relative error is assumed, gave the overall best
`fit. The choice of model was made with respect to resid-
`ual plots and parameters precision. A lag-time (tlag) was
`included
`for oral data. The volume of distribution at
`steady state (V,,), systemic clearance (CL), bioavailabil-
`ity (F), and elimination half-life
`(txp) were estimated
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 3
`
`

`
`532
`
`Brynne et al.
`
`CLINICAL
`
`PHARMACOLOGY
`
`& THERAPEUTICS
`MAY 1998
`
`Oral administration
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`12
`Intravenous
`
`0
`infusion
`1000
`
`1
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`
`0
`
`2
`
`4
`
`6
`Time
`
`(h)
`
`8
`
`10
`
`12
`
`0
`
`2
`
`4
`
`8
`
`10
`
`12
`
`6
`Time
`
`(h)
`
`Fig. 2. Serum concentration-time profiles of tolterodine and the S-hydroxymethyl metabolite
`(5HM) after oral administration of 4 mg tolterodine tartrate twice a day (top panel) and intra-
`venous infusion of 1.8 mg tolterodine tartrate (bottom panel)
`for extensive metabolizers (n = 8)
`and poor metaholizers (n = 8).
`
`according to standard equations.l7J* Noncompartmen-
`tal analysis of tolterodine and 5-HM data was performed
`to determine the metabolite-parent
`compound ratio for
`area under the serum concentration-time
`curve (AUC)
`and terminal half-life
`(tXZ) of 5-HM. The AUC was
`obtained by means of linear trapezoidal
`rule17 with
`extrapolation to infinity by means of division of the last
`calculated data point by the terminal slope (A,) derived
`from the 2- to 24-hour interval. Statistical analysis for
`pharmacokinetic variables was performed with the Stu-
`dent t test for unpaired data. Differences were consid-
`ered to be significant at p < 0.05.
`The pharmacologic effect of tolterodine on salivation
`was expressed as the area under the effect curve for the
`first 8 hours (.AUEC) and computed according to the lin-
`
`relation,
`ear trapezoidal rule. In the concentration-effect
`the effect was transformed by use of the relative change
`from the baseline values computed as follows:
`
`E _ Ei
`
`- EO,i
`
`EO,i
`
`in which Ei is the value of the salivary effect at time i,
`and Eo,i is the corresponding value at baseline.
`
`RESULTS
`Phmmucokinetics. The individual serum concentratio*
`time profiles of tolterodine and 5-HM at steady state
`(oral) and after intravenous infusion are shown in Fig 2.
`There was a distinct difference in serum tolterodine con-
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 4
`
`

`
`CLINICAL
`VOLUME
`
`PHABMA COLOGY & THERAPEUTICS
`63, NUMBER 5
`
`Brynne et al. 533
`
`Table I. Pharmacokinetic parameters of tolterodine after simultaneous fit of oral multiple-dose data (4 mg twice
`daily) and intravenous infusion data (1.8 mg) after administration of tolterodine
`to extensive and poor metabolizers
`CL (Zh)
`Volunteer No.
`
`flag (hr)
`
`t,,,, (hr)
`
`Cm, WOW
`
`F f%)
`
`vss 69
`
`txp (W
`
`Extensive metabolizers
`1
`2
`3
`4
`11
`12
`15
`16
`Mean
`SD
`Poor metabolizers
`5
`6
`7
`8
`9
`10
`13
`14
`Mean
`SD
`
`0.36
`0.63
`0.27
`1.0
`0.49
`0.72
`0.49
`0.10
`0.51
`0.28
`
`0.30
`0.30
`0.31
`0.30
`0.33
`0.75
`0.29
`0.74
`0.42
`0.20
`
`per liter.
`Values in parentheses are micrograms
`CL, systemic
`clearance;
`c max, peak serum concentration;
`distribution
`at steady state.
`*p < 0.001 versus extensive metabolizers.
`tp < 0.002 versus extensive metabolizers.
`
`1.0
`1.0
`0.5
`2.0
`1.5
`1.5
`1.0
`1.0
`1.2
`0.5
`
`2.0
`2.0
`4.0
`1.0
`1.0
`2.0
`1.0
`2.0
`1.9
`1.0
`
`18 (5.9)
`1.9 (0.63)
`43 (14)
`5.5 (1.8)
`8.3 (2.7)
`4.9 (1.6)
`43 (14)
`2.8 (0.90)
`16 (5.2)
`17 (5.7)
`
`71 (23)
`123 (40)
`89 (29)
`101 (33)
`212 (69)
`117 (38)
`141 (46)
`71 (23)
`116* (38)
`46 (15)
`
`21
`40
`58
`11
`13
`15
`40
`6.8
`26
`18
`
`89
`99
`48
`75
`170
`77
`120
`49
`91-t
`40
`
`89
`158
`135
`134
`101
`154
`90
`147
`126
`28
`
`181
`110
`102
`111
`104
`87
`118
`110
`115
`28
`
`35
`53
`33
`40
`41
`69
`31
`52
`44
`13
`
`13
`8.1
`9.7
`7.7
`8.9
`5.9
`10
`8.6
`9.0*
`2.1
`
`1.9
`2.1
`3.6
`2.4
`1.9
`1.8
`2.6
`2.4
`2.3
`0.6
`
`9.9
`10
`7.5
`10
`8.2
`11
`8.3
`9.0
`9.2*
`1.2
`
`F, absolute bioavailability;
`
`tlagr lag-time;
`
`t,,,=.
`
`time to reach C,,;
`
`tYzp. elimination half-life; Vss, volume of
`
`Table II. Pharmacokinetic values of 5-HM after oral administration
`tolterodine among extensive metabolizers
`
`(4 mg twice daily) and metabolic ratios of
`
`Volunteer No.
`
`tmnx W
`
`C,,
`
`(nmoWL)
`
`txz W
`
`1
`2
`3
`4
`11
`12
`15
`16
`Mean
`SD
`
`1.0
`1.5
`0.5
`2.0
`1.5
`1.5
`1.0
`0.5
`1.2
`0.5
`
`15 (5.2)
`7.3 (2.5)
`25 (8.6)
`9.1 (3.1)
`11 (3.7)
`12 (4.1)
`27 (9.1)
`7.0 (2.4)
`14 (4.8)
`7.7 (2.6)
`
`3.0
`2.5
`3.0
`3.0
`2.7
`2.5
`3.7
`2.7
`2.9
`0.4
`
`Single
`dose
`
`0.84
`ND
`0.94
`0.51
`0.68
`0.50
`1.1
`0.16
`0.59
`0.38
`
`Steady
`state
`
`0.78
`0.24
`0.99
`0.49
`0.59
`0.37
`0.89
`0.32
`0.58
`0.28
`
`Infusion
`
`3.3
`3.4
`3.4
`ND
`4.4
`2.6
`2.3
`3.2
`3.3
`0.67
`
`per liter.
`in parentheses are micrograms
`Values
`time to reach C,,,; C,,
`curve; hax.
`AUC, Area under
`the serum concentration-time
`slope (&) of the semilogarithmic
`serum concentration-time
`curve; ND, not determined.
`
`peak serum concentration;
`
`tYIz, terminal half-life
`
`associated with
`
`the terminal
`
`centrations between the panels of the extensive metabo-
`lizers and those of the poor metabolizers. The 5-HM lev-
`els for extensive metabolizers were similar to those of
`tolterodine, but 5-HM was not quantifiable among poor
`
`metabolizers. The pharmacokinetic parameters of toltero-
`dine are given in Table I. Tolterodine was rapidly
`absorbed. Absorption half-life was 0.41 f 0.23 hour
`among extensive metabolizers and 0.53 + 0.40 hour
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 5
`
`

`
`534
`
`Brynne et al.
`
`CLINICAL
`
`PHARMACOLOGY
`
`& THERAPEUTICS
`MAY 1998
`
`Extensive
`
`metabolizers
`
`Poor metabolizers
`
`01
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`
`! , , , , , , ,
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`
`Time(h)
`
`Time(h)
`
`Fig. 3. Mean + SEM heart rate, near point of vision, and salivation versus time profiles at base-
`line (open circles) and after oral administration of 4 mg
`twice a day (solid circles) and intravenous
`infusion of 1.8 mg tolterodine
`(squares) among eight extensive metabolizers
`(left panel) and eight
`poor metabolizers
`(right panel).
`*Five of eight poor metabolizers were accidentally given lunch
`before the measurement. The y-axes of heart rate and near point of vision have been cut off.
`
`among poor metabolizers. There was a slight tendency
`for increased time
`to maximum serum levels (t,,)
`among poor metabolizers. A significant sevenfold differ-
`ence (p < 0.001) in peak serum tolterodine concentration
`(C,,,) was observed between the two groups at steady
`state. The initial phase was rapid (tXcc < 0.2 hour) in all
`subjects, with the exception of volunteer 6 (0.8 hour).
`Three subjects (3, 9, and 13) exhibited very high F val-
`ues compared with the average. There was a highly sig-
`nificant difference (p < 0.001) in CL between the panels
`and therefore txp was fourfold longer among poor metab-
`olizers @ < 0.001). The blood/serum concentration ratio
`of tolterodine was similar for both groups (0.60 f 0.18
`and 0.57 + 0.05, respectively). Excretion of unchanged
`drug in urine (P-glucuronidase-treated samples) was less
`than 2.5% for all subjects.
`The pharmacokinetic values of 5-HM among exten-
`sive metabolizers are given in Table II. The C,,, of the
`metabolite was obtained at the same t,,, as the parent
`
`(t& was slightly
`compound, but the terminal half-life
`longer, 2.9 f 0.4 hours. Although serum levels of 5-HM
`were quantifiable after intravenous administration, half-
`life was difficult to estimate (because of the limit of quan-
`tification). The molar AUCtolterodine to AUC,-,
`oral ratio
`was time independent (single versus multiple dose) and
`one-sixth of the intravenous value. About 5% of the
`administered tolterodine dose was excreted as 5-HM in
`urine. Although
`the metabolite was quantifiable
`in the
`urine of poor metabolizers, the excreted fraction was less
`than 1% of the administered dose.
`relation. The
`Pharmacokinetic-pharmacodynamic
`effect of tolterodine on heart rate, near point of vision,
`and salivation
`is shown in Fig. 3. Among extensive
`metabolizers, a slight increase in heart rate (10 f 8.4
`beats/mm) was seen at steady state compared with base-
`line. In contrast, no effect was evident among poor
`metabolizers except a decrease after the intravenous
`infusion. A slight increase in accommodation
`(41 mm)
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 6
`
`

`
`CLINICAL
`VOLUME
`
`I’ HARMACOLOGY
`63, NUMBER 5
`
`& THERAPEUTICS
`
`Brynne et al.
`
`535
`
`18-
`l8 _ Extensive
`
`metabolizers
`
`Oral administration
`’
`18-
`,6 _ Poor metabolizers
`
`e+
`/’
`A,’
`I’
`
`I’
`
`I’
`
`I’
`
`,’
`
`/’
`
`14-
`
`12-
`
`lo-
`
`I’
`
`/’
`
`/’
`
`+x’
`
`q/’
`I
`
`0
`
`I’
`
`/’
`
`I’
`
`/’
`
`14-
`
`12-
`
`lo-
`
`8-
`
`8-
`
`18
`
`16
`
`0
`
`2
`
`4
`
`8
`
`8
`
`10
`
`12
`
`18
`18
`14
`Intravenous
`
`2
`0
`infusion
`
`4
`
`6
`
`8
`
`10
`
`12
`
`14
`
`16
`
`18
`
`0
`
`16
`
`18
`
`0
`
`14
`4
`2
`8
`6
`12
`10
`8
`6
`14
`12
`10
`4
`2
`Salivation
`AUEC
`(O-8 h) at baseline
`Salivation
`AUEC
`(O-8 h) at baseline
`Fig. 4. Relation between area under the effect curve (AUEC) for salivation (0 to 8 hours) at steady
`state versus baseline after 4 mg tolterodine tartrate twice a day (top panel) and during intravenous
`infusion versus baseline of 1.8 mg tolterodine tartrate (bottom panel) among extensive metaboliz-
`ers (left graphs) and poor metabolizers (right graphs).
`
`16
`
`18
`
`was evident among poor metabolizers, but the other
`volunteers remained unaffected. Basal salivation did
`not drift adversely in the two groups except for the 4-
`hour value among poor metabolizers, which was higher
`than expected because five of the eight subjects were
`accidentally given lunch before the measurement.
`Tolterodine caused a decrease in salivation among
`all subjects; a maximum effect was obtained within
`2 hours of administration. Because of the large differ-
`ence in pharmacokinetics
`between
`the groups,
`the
`effect of tolterodine on basal saliva secretion was ana-
`lyzed with respect to the dose given. Fig. 4 illustrates
`the relation between saliva AUEC(O-8) for oral steady
`state versus baseline and intravenous
`infusion versus
`
`the two phenotypes.
`baseline among subjects from
`There was pronounced
`interindividual
`variation,
`within and between
`the two groups, in basal saliva
`secretion. A distinct drug effect was nevertheless
`obtained for four of eight extensive metabolizers and
`most of the poor metabolizers after oral administra-
`tion. For extensive metabolizers the effect was equally
`pronounced
`after
`intravenous
`compared with oral
`administration, whereas salivation was less affected
`among poor metabolizers after the infusion.
`In previous studies, it was concluded that the effect
`on stimulated salivation after tolterodine administra-
`tion was mainly derived
`from an active unknown
`metabolite”
`and that there was a tenfold difference in
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 7
`
`

`
`536 Brynne et al.
`
`CLINICAL
`
`PHARMA COLOGY & THERAPEUTICS
`MAY 1998
`
`A
`
`0.2,
`
`-0.6
`
`1
`
`-0.6
`
`/q
`
`Extensive
`
`metabolizers
`
`Extensive
`
`metabolizers
`
`B
`
`0.2
`
`l
`
`’
`0
`
`l
`CfOe
`
`0
`
`5
`
`l
`
`0
`
`0
`
`0
`
`-0.6
`
`-0.6 ~/q
`
`1
`1
`0.1
`(nM)
`concentration
`tolterodine
`Unbound
`(nM)
`concentration
`5-HM
`Unbound
`Fig. 5. Median salivation effect (0 to 8 hours) versus unbound serum tolterodine (A) and 5-HM
`(B) concentration after single and multiple oral doses (solid circles) and during intravenous infu-
`sion (open circles) of tolterodine among extensive metabolizers and poor metabolizers.
`
`10
`
`10
`
`serum protein binding between tolterodine and 5-HM
`(fraction unbound
`[f,,] of tolterodine, 3.7%, and f, of
`5-HM, 36%).*9 To explain the similarity
`in salivary
`effects between
`the
`two phenotypic
`groups,
`the
`unbound serum concentrations of tolterodine, 5-HM,
`and the combination of tolterodine and 5-HM were
`studied. The relation between salivary effect (0 to 8
`hours) and unbound serum concentrations of toltero-
`dine and 5-HM for extensive metabolizers is shown in
`Fig. 5, A. There was a weak correlation between toltero-
`dine concentration and effect on salivation. A stronger
`correlation was seen with 5-HM concentration and
`effect
`(Fig. 5, B). The strongest correlation was
`obtained when the concentrations of both tolterodine and
`5-HM were taken into account (assuming no competi-
`tive antagonistic
`interaction
`[slope = 1.03; r-2 = 0.621;
`Fig. 6, A). A correlation also was apparent between
`unbound serum tolterodine concentration and effect on
`salivation (Fig. 6, B) among poor metabolizers (slope =
`1.14; ~-2 = 0.50), irrespective of route of administration
`or duration of oral administration
`(i.e., single dose or
`steady state). No apparent hysteresis was observed for
`the serum concentration versus salivation data.
`All 16 volunteers completed
`the study. No severe
`adverse events were reported. The most frequently
`reported adverse events were headache (two extensive
`metabolizers and four poor metabolizers), dry mouth
`(four extensive metabolizers and two poor metaboliz-
`ers), abnormal visual accommodation (five poor metab-
`olizers), and tachycardia (four extensive metabolizers).
`
`Most events were judged as mild. There were no clini-
`cally significant changes in blood chemistry, blood
`pressure, or electrocardiographic values (QT, or mor-
`phology) among the volunteers during the study.
`
`DISCUSSION
`is extensively
`This study showed that tolterodine
`metabolized by CYP2D6. The high specificity is shown
`by the fivefold difference in CL between the two pan-
`els and the fact that poor metabolizers showed no quan-
`tifiable serum levels of 5-HM.
`If it is assumed that
`enzyme capacity is the only difference between the pan-
`els, the findings imply that at least 80% of a systemi-
`cally available dose of tolterodine
`is metabolized by
`CYP2D6
`to 5-HM
`in extensive metabolizers. The
`approximately
`sixfold higher AUC
`ratio (5-HM
`to
`tolterodine) between oral and intravenous administra-
`tion suggests that at least 85% of the biotransformation
`of tolterodine occurs during first pass. If the CL values
`are adjusted for the blood to serum concentration ratio
`and if a hepatic blood flow of 1.5 Wmin
`is assumed,
`tolterodine
`is a high-extraction drug (81%) for exten-
`sive metabolizers and a low-extraction drug (18%) for
`poor metabolizers. Compared with these estimates, the
`obtained values of F were higher in both panels. How-
`ever, three volunteers displayed very high F values of
`170%, 120% and 58%. Because tolterodine
`is highly
`bound to serum proteins (f,, 3.7%), primarily to al-acid
`glycoprotein,tg
`this can result in intraindividual vari-
`ability caused by fluctuations
`in al-acid glycoprotein
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 8
`
`

`
`CLINICAL
`VOLUME
`
`P HARMACOLOGY
`63, NUMBER 5
`
`& THERAPEUTICS
`
`Brynne et al.
`
`537
`
`A
`
`0.2 1
`
`0.0
`
`Extensive
`metabolizers
`
`‘\
`
`‘\\
`
`“&
`
`‘\
`
`B
`
`0.2 1 Poor I metabolizes
`
`0.0
`
`-0.2
`
`-0.4
`
`-
`
`-0.6
`
`-
`
`1
`
`-0.6
`
`!
`
`10
`(nM)
`concentration
`tolterodine
`Unbound
`(nM)
`concentration
`toRerodine+SHM
`Unbound
`Fig. 6. Regression analysis and 95% confidence intervals of the median salivation effect (0 to 8
`hours) versus unbound serum tolterodine with 5-HM (A) and tolterodine (B) concentration after
`single and multiple oral doses (solid
`and during intravenous infusion (open
`of
`circles)
`circles)
`tolterodine among extensive metabolizers and poor metabolizers.
`
`‘\
`
`I
`1
`
`I
`10
`
`levels increased
`levels. Indeed, al-acid glycoprotein
`(50% to 100%) in all three volunteers after oral admin-
`istration compared with
`intravenous
`infusion, which
`probably explains the deviation in F values.
`Regional selectivity of oxidation
`is consistent for
`almost all CYP2D6 substrates. Oxidation occurs in the
`hydrophobic region 5 to 7 A from the basic nitrogen.20221
`In an x-ray crystallographic analysis of tolterodine,
`the
`corresponding distance was determined
`to be 5.78 A
`(Engstrbm I, Tergenius LE, Sellberg B. Personal com-
`munication, 1997). Another common feature of many
`CYP2D6 substrates is that interphenotype differences in
`metabolism become less prominent at steady state than
`after single doses, presumably because of enzyme satu-
`ration. This has been reported for several drugs, includ-
`ing clomipramine,** paroxetine,*j propafenone,*4 and
`methotrimeprazine
`(INN, levomepromazine).*s Toltero-
`dine has shown dose proportionality
`in single doses up
`to 12.8 rng.” Despite the high CYP2D6 specificity, no
`apparent saturation was observed after multiple doses as
`determined according to metabolic ratio. Furthermore,
`the decline in concentration of 5-HM was slightly slower
`than that of the parent compound and was not affected
`during twice-a-day administration.
`Only minute amounts of parent compound were
`recovered in both urine and feces in a mass-balance
`study. 1 1 Metabolism
`is thus also the main route of elim-
`ination among poor metabolizers because less than 2.5%
`of intact tolterodine was excreted in urine. In an in vitro
`
`study, hydroxylation of tolterodine showed strong cor-
`relation with CYP2D6 activity, whereas dealkylation
`correlated with CYP3A activity.*6 Taken together, these
`findings imply that tolterodine
`is eliminated by at least
`two parallel pathways: a high-affinity,
`low-capacity
`pathway
`(hydroxylation
`by CYP2D6)
`in extensive
`metabolizers and a low-affinity, high-capacity pathway
`(dealkylation by CYP3A) among poor metabolizers.
`In contrast to the kinetic data, the pharmacodynam-
`its of tolterodine were not generally
`influenced by
`metabolic phenotype.
`In a previous study, oral single-
`doses of tolterodine produced a distinct
`increase in
`heart rate at 6.4 mg, whereas visual accommodation
`was only affected at 12.8 mg.”
`In the present study, at
`a dosage of 4 mg, only a small increase in heart rate
`was observed in extensive metabolizers, whereas heart
`rate was unaffected among poor metabolizers, although
`the near point of visual accommodation
`increased. The
`effect on salivation was more distinct, without any
`major differences between the panels. This shows that
`F (26% in extensive metabolizers and 9 1% among poor
`metabolizers)
`is not a useful parameter to explain the
`pharmacologic effect of tolterodine.
`the salivary effect
`Among extensive metabolizers,
`could not be explained with tolterodine concentrations
`alone. Although depressed saliva secretion and 5-HM
`concentration
`resulted in a stronger relation after oral
`administration,
`they did not accurately explain
`the
`more complex situation after intravenous administra-
`
`Petitioner Mylan Pharmaceuticals Inc. - Exhibit 1011 - Page 9
`
`

`
`538
`
`Brynne et al.
`
`CLINICAL
`
`P HARMACOLOGY
`
`& THERAPEUTICS
`MAY 1998
`
`tion. If equipotency of the two compounds12 and addi-
`tivity are assumed, the sum of the unbound
`toltero-
`dine and 5-HM concentration
`correlated with
`the
`effect on salivation
`independent of route of adminis-
`tration among extensive metabolizers. This suggests
`that during
`infusion
`the effect
`is mainly caused by
`tolterodine, whereas after cessation of infusion
`the
`effect can be attributed
`to increasing levels of 5-HM.
`Indeed, 30 minutes after the infusion most of the effect
`on salivation was related
`to 5-HM. Among poor
`metabolizers
`there was a correlation between toltero-
`dine concentration and the effect on salivation
`irre-
`spective of the route of administration.
`The two concentration-effect
`relations suggest that
`5-HM
`is the only major pharmacologically
`active
`metabolite after tolterodine administration because oral
`and intravenous data correlate well in both panels. Our
`data indicate an unbound 50% inhibitory concentration
`on salivation of about C,,,,
`that is, 3 to 4 nmol/L
`in
`extensive metabolizers and poor metabolizers. These
`estimates are very similar to the inhibition constants for
`parotid glands in vitro (4.8 nmol/L
`for tolterodine and
`5.2 nmol/L
`for 5-HM).l* The C max of total tolterodine
`(bound and unbound) among extensive metabolizers
`(16 nmol/L, or 5.2 yg/L) correlates well with the sug-
`gested 50% inhibitory concentration of 6 to 8 pg/L for
`stimulated
`salivation.11 Consequently,
`a true 50%
`inhibitory concentration
`for saliva d