carvedilol pharmacokinetics and pharmacodynamics in relation to

ReseaRch aRticle

ISSN 1462-2416Pharmacogenomics (2011) 12(6), 783–79510.2217/PGS.11.20 © 2011 D Sehrt, I Meineke, M Tzetkov, S Gültepe & J Brockmöller

783

Carvedilol pharmacokinetics and pharmacodynamics in relation to CYP2D6 and ADRB pharmacogenetics

BackgroundThe b-blocker carvedilol is effective against hypertension and can prolong the survival of patients with chronic heart failure. Carvedilol is administered as a racemic drug. The S-enantiomer mediates b

1 and b

2 adreno receptor

blockade, while both enantiomers mediate a1

adrenoreceptor blockade [1]. Carvedilol metab-olism is mediated by CYP2D6, with possible contributions by CYP1A2, CYP2C9, CYP2C19 and CYP3A4 [2]. Inherited CYP2D6 polymor-phisms have been shown to cause variation in carvedilol pharmaco kinetics [3–7]. Comparing six extensive and five poor metabolizers of CYP2D6 substrates, greater blood pressure reduction was found in the poor metabolizer group [6]. By con-trast, inhibition of CYP2D6 by fluoxetine did not significantly alter cardiovascular parameters in ten patients suffering from heart failure [8]. Furthermore, clinically adjusted carvedilol doses in a study of 74 patients with chronic heart fail-ure were not lower but higher in poor metaboliz-ers than in extensive metabolizers [9] but no blood concentration data were available in this study.

In addition to these controversial data on the pharmacodynamic effects of the poor versus extensive CYP2D6 genotypes, only scarce data

are available on the differential consequences of the frequent CYP2D6 alleles *1, *2, *9, *17, *35 and *41 on carvedilol pharmacokinetics. Most of these different CYP2D6 alleles are coding for CYP2D6 protein variants (Supplementary table 1 for details; www.futuremedicine.com/doi/suppl/10.2217/pgs.11.20), which may well have differential activities with different substrates. The primary aim of the clinical trial presented here was to quantify the effects of CYP2D6 poly-morphisms on carvedilol pharmacokinetics and to measure how these pharmacokinetic differ-ences correlate with carvedilol-induced reduc-tion of resting and exercise-induced heart rate and with carvedilol effects on blood pressure.

Response to carvedilol may also be modu-lated by genetic variation in b

1 adrenoreceptors

(ADRB1). In particular, the ADRB1 variants Ser49Gly and Arg389Gly have been character-ized in molecular and clinical studies [10–15]. The ADRB1 Arg389Gly polymorphism appears most interesting since real-time optical recording has demonstrated dramatic differences in response to carvedilol between the two codon

389 alleles [16].

For analysis of clinical study data it is relevant to know that the codon

49 and codon

389 variants are

found in three haplotypes, Ser49Arg389 (H1),

Aims: Carvedilol is an effective treatment in hypertension and chronic heart failure. The medical impact of polymorphisms in CYP2D6 and in the b-adrenergic receptors ADRB1 and ADRB2 on the pharmacokinetics and pharmacodynamics of carvedilol is controversial. Methods: After carvedilol 25 mg was administered to 110 volunteers, concentrations were enantioselectively quantified and effects on resting and exercise-induced heart rate and blood pressure were analyzed using population pharmacokinetic, pharmacodynamic and pharmacogenetic modeling. Results: There were significant CYP2D6 allele-specific differences in carvedilol pharmacokinetics, but the CYP2D6 genotype had no effect on heart rate, blood pressure or adverse effects. ADRB1 Gly49 was associated with higher baseline heart rates and with greater carvedilol effects on exercise heart rates. Carriers of ADRB2 Gln27 had greater reduction in resting blood pressure by carvedilol compared with Glu27. Conclusion: Carvedilol is a drug where CYP2D6-related pharmacokinetic variation is apparently not carried forward into pharmacodynamic variation. Although current knowledge does not allow utilizing ADRB1 and ADRB2 genotypes for clinical treatment decisions, our data should stimulate further research on the impact of these genotypes in health and disease.

Original submitted 16 November 2010; Revision submitted 19 January 2011

KEYWORDS: ADRB1 n ADRB2 n adrenergic receptor polymorphism n carvedilol n CYP2D6 polymorphism n heart failure n hypertension n PKPD modeling n population pharmacokinetics

Daniel Sehrt1, Ingolf Meineke1, Mladen Tzvetkov1, Senol Gültepe1 & Jürgen Brockmöller†1

1University Medicine Göttingen, Department of Clinical Pharmacology, Robert-Koch-Str. 40, D-37075 Göttingen, Germany †Author for correspondence:Tel.: +49 551 395 770 Fax: +49 551 391 2767 [email protected]

For reprint orders, please contact: [email protected]

ReseaRch aRticle Sehrt, Meineke, Tzvetkov, Gültepe & Brockmöller

future science groupPharmacogenomics (2011) 12(6)784

Ser49Gly389 (H2) and Gly49Arg389 (H3). Earlier, haplotype H1 was associated with poor outcome of antihypertensive therapy [17,18]. The ADRB1 Arg389Gly polymorphism was associ-ated with the dose of carvedilol administered in heart failure [9].

Carvedilol may exert some of its clinical effects via binding to the b

2 adrenoreceptor (ADRB2)

[19] and besides several rare variants there are two frequent functional polymorphisms in ADRB2, Arg16Gly and Gln27Glu, with a controversial impact in carvedilol therapy [14,20,21]. Thus, it was the intention of the present study to confirm and quantify the impact of CYP2D6, ADRB1 and ADRB2 genotypes on carvedilol pharmacokinetics and pharmacodynamics. The present clinical phar-macokinetic/pharmacodynamic/pharmacogenetic (PKPDPG) analysis should allow differentiating between the contribution of pharmaco kinetic vari-ation (CYP2D6) and pharmacodynamic variation (ADRB1 and ADRB2) and should thus contribute to a better understanding of these variants in the healthy human being.

Patients & methodsA detailed explanation of the clinical study and of the bioanalytical methods is available at the website of the journal www.futuremedicine.com/doi/suppl/10.2217/pgs.11.20.

ResultsThe carvedilol PKPDPG analysis was performed for 36 female and 74 male healthy volunteers with a median (range) age of 28 (20–48) years, bodyweight of 72 (52–99) kg and BMI of 23.2 (17.4–29.8) kg/m2. An extensive vari-ation in pharmacokinetics was found (FigureS 1

& 2). Maximum carvedilol plasma concentra-tions measured 1 h after administration varied nearly 14-fold with a median (range) of 99.3 (24.0–327.2) nmol/l for R-carvedilol and 32.5 (7.9–109) nmol/l for S-carvedilol. The area under the concentration time curve extrapo-lated to infinity (AUC

0-infinity) of R-carvedilol

varied 26-fold with 363 (111–2922) nmol*h/l, and that of S-carvedilol varied 14-fold with 200 (58.9–895.5) nmol*h/l. Total clearances increased with increasing number of CYP2D6 active genes, but by classifying genotypes into five groups, linear regression analysis showed that CYP2D6 genotype only accounted for 24.4 and 8.2% of the variation in total clearance of R- and S-carvedilol, respectively (table 1).

Maximum concentrations and AUCs of the active metabolite desmethylcarvedilol decreased in parallel to the decreasing concentration of the parent drug with increasing number of active CYP2D6 genes (tableS 1 & 2). By contrast, maxi-mum plasma concentrations (C

max) of both

500.0400.0300.0200.0

100.070.050.040.030.020.0

10.07.05.04.03.02.0

1.00.70.50.40.30.2

0.10 12 24 36 48

Time (h) Time (h)

Co

nce

ntr

atio

n (

nm

ol/l

)

0 12 24 36 48

S-carvedilolR-carvedilol

Figure 1. Interindividual carvedilol pharmacokinetic variation. Individual measurements of the plasma concentrations after administration of 25 mg racemic carvedilol (12.5 mg = 30.75 µmol per enantiomer) are illustrated by the black lines. Red curves are population mean simulations for typical subjects with the genotypes CYP2D6 *1/*1 (lowest), CYP2D6 *1/0 (middle) and CYP2D6 0/0 (upper), with the mean bodyweight of the sample as 73 kg. As seen, at all time points there is a significant more than tenfold (range) variation in pharmacokinetics of both carvedilol enantiomers, which was only partially explained by CYP2D6 genotype and bodyweight.

Carvedilol: pharmacokinetics/pharmacodynamics/pharmacogenetics ReseaRch aRticle

www.futuremedicine.comfuture science group 785

enantiomers of 4 -́OH-carvedilol did increase with increasing number of active CYP2D6 genes (p < 0.003 [table 2]) and the AUCs showed a cor-responding trend. There were no such significant differences in AUC or C

max of 5́ -OH-carvedilol

depending on CYP2D6 (tableS 1 & 2).Population pharmacokinetic modeling

revealed a quantitatively relevant CYP2D6 inde-pendent clearance (Cl

0), with a mean of 27.1 l/h

for R-carvedilol (Cl/F 29.4 l/h) and 51.9 l/h for S-carvedilol (Cl/F 107.5 l/h). In homozygous extensive metabolizers of CYP2D6 substrates (CYP2D6*1/*1), 74% of R-carvedilol and 50% of S-carvedilol was eliminated via CYP2D6 as calculated from the difference between extensive and poor metabolizers (tableS 3, 4 & 5). The low allele-specific clearances of S-carvedilol mediated by CYP2D6*9, *10, *17 and *41 could not be distinguished statistically (tableS 3, 4 & 5). In con-trast to our expectations, R-carvedilol clearances mediated by the alleles CYP2D6*2 and *35 were smaller than those by CYP2D6*1 as expressed by CYP2D6 allele-specific R-carvedilol clearances

of 5.9, 8.3 and 12.9 l/h for *35, *2 and *1 (tableS 4

& 5). This result is also illustrated by compari-son of genotypes such as CYP2D6*2/0 and CYP2D6*35/0 versus CYP2D6*1/0, where zero denotes any completely deficient allele (Figure 2). When considering all possibly differentially active CYP2D6 alleles, the CYP2D6 genotype still only explained 35.7 and 15.3% of the vari-ation in total clearance of R- and S-carvedilol, respectively. However, CYP2D6 genotype was the most important predictor in the present study. Bodyweight only explained 1.6% (not significant) and 4.2% of total oral clearance of R- and S-carvedilol. Sex explained only 0.6% (not significant) and 5% of the variation in total clearance of R- and S-carvedilol, respectively.

With the limitations due to the method of drug administration (only oral dosing), pharmaco-kinetic modeling supported a significant contri-bution of CYP2D6 both to first-pass metabolism of carvedilol and to systemic clearance. Estimates for the CYP2D6-dependent first-pass metabolism are illustrated in Supplementary table 3.

160

120

80

40

0

R-c

arve

dilo

l to

tal c

lear

ance

(C

l/F; l

/h)

0/0

0/*1

7*1

7/*1

70/

*41

*17/

*41

0/*3

50/

*90/

*10

*41/

*41

0/*2

*9/*

41*3

5/*3

5*9

/*35

0/*1

*10/

*10

*2/*

35*2

/*9

*2/*

10*2

/*2

*1/*

41*1

/*35

*1/*

9*1

/*10

*1/*

2*1

/*1

CYP2D6 genotype

0S

-car

ved

ilol t

ota

l cle

aran

ce (

Cl/F

; l/h

)

0/0

0/*1

7*1

7/*1

70/

*41

*17/

*41

0/*3

50/

*90/

*10

*41/

*41

0/*2

*9/*

41*3

5/*3

5*9

/*35

0/*1

*10/

*10

*2/*

35*2

/*9

*2/*

10*2

/*2

*1/*

41*1

/*35

*1/*

9*1

/*10

*1/*

2*1

/*1

CYP2D6 genotype

400

350

300

250

200

150

50

100

Figure 2. Genotype–phenotype correlation between CYP2D6 and R- and S-carvedilol oral total clearance. Individual R- and S-carvedilol oral total clearances are shown as circles and population mean predictions for genotypes are shown as horizontal bars. These mean clearances depicted by the horizontal bars are those normalized for 73 kg bodyweight, while the individual estimates depicted here are not normalized for bodyweight, which explains some but not all variation within the CYP2D6 genotype groups. CYP2D6 alleles *3, *4, *5, and *6 are not differentiated further and are labeled as 0. As explained in the supplementary methods section, clearance of any subject can be predicted as the sum of a CYP2D6-independent clearance and two CYP2D6 allele-specific clearances. The order of the genotypes in both graphs is by increasing R-carvedilol clearance.



CYP2C9 and CYP2C19 may also contribute to carvedilol metabolism. In our study sam-ple there were 77, 16, 3, 12 and 2 carriers of CYP2C9 *1/*1, *1/*2, *2/*2, *1/*3 and *2/*3, respectively, which allowed a reasonably power-ful analysis of the impact of CYP2C9 polymor-phisms on the pharmacokinetics of carvedilol. However, there were no significant differences in the C

max and AUC of carvedilol and its metabo-

lites. There were 8, 23, 40, 10, 21 and 8 carriers of CYP2C19 *17/*17, *1/*17, *1/*1, *2/*17, *1/*2 and *2/*2, respectively, but there were also no differences in the pharmacokinetics of carvedilol and its metabolites.

n Pharmacodynamics: heart rateThe resting heart rates prior to carvedilol (Supplementary table 4) were independent from CYP2C9, -2C19, -2D6 and ADRB2 genotypes,

but significantly different among the ADRB1 genotypes (FigureS 3 & 4). Stratified and multi-factorial analyses, including age, gender, body-weight, body height and ethnicity, demonstrated that the ADRB1-related difference was not due to confounding by the other factors. Resting heart rates were higher with increasing number of Gly49 alleles (mean of 81, 87, and 89 bpm in carriers of 0, 1 and 2 Gly alleles; n = 75, 31 and 4, respectively).

Maximum heart rate reduction after carvedilol (Supplementary table 4) was not correlated with gen-der, age, body height, CYP genotypes or ADRB genotypes. In addition, the heart rate reduction integrated over the first 24 h after carvedilol administration was not significantly correlated with gender, age, body height, CYP genotype or ADRB genotype. However, maximum and 24-h integrated heart rate reduction by carvedilol did

Table 2. Maximum plasma concentrations of R- and S-carvedilol and their metabolites in relation to CYP2D6 genotype.

Substance Median (interquartile ranges [q1–q3] [nmol•h/l]) p-value (trend)‡

Number of active CYP2D6 genes†

0 (n = 13) 0.5 (n = 12) 1 (n = 22) 1.5 (n = 14) ≥2 (n = 38)

R-carvedilol 154 (140–223) 89.2 (55.8–122) 123 (80.3–148) 99.2 (39.4–135) 72.4 (52.2–102) <0.0001S-carvedilol 43.3 (35.1–59.2) 24.2 (13.8–31.0) 36.3 (28.2–50.8) 36.6 (18.6–45.9) 27.8 (22.3–37.0) 0.07R-desmethylcarvedilol 8.1 (6.1–21.9) 5.6 (2.8–9.2) 9.9 (5.7–15.4) 6.1 (4.1–11.1) 4.7 (2.8–8.3) <0.0001S-desmethylcarvedilol 17.0 (13.8–22.2) 11.2 (5.8–18.5) 11.3 (8.8–16.9) 10.2 (6.6–15.1) 7.5 (5.3–13.4) 0.005R-4’-OH-carvedilol 3.95 (3.43–5.38) 5.02 (2.63–6.19) 4.98 (3.94–7.70) 6.35 (5.02–9.24) 7.26 (5.71–9.00) <0.0001S-4’-OH-carvedilol 10.5 (5.8–13.4) 8.9 (4.0–11.4) 10.0 (7.8–16.6) 13.4 (9.0–19.4) 12.8 (9.6–18.6) 0.003R-5’-OH-carvedilol 12.2 (6.0–14.8) 10.6 (4.9–13.5) 9.5 (6.5–12.4) 9.3 (7.3–13.2) 8.4 (7.2–12.3) NSS-5’-OH-carvedilol 12.4 (6.7–15.6) 9.8 (6.6–14.3) 10.0 (7.7–12.9) 9.9 (7.5–13.6) 9.5 (7.7–13.4) NS†For this summarizing analysis, 0.0 activity units were attributed to CYP2D6 alleles *3, *4, *5, *6 and *7, 0.5 activity units were attributed to alleles *9, *10, *17 and *41 and 1.0 activity unit was attributed to *1, *2 and *35. There was only one ultrarapid metabolizer in this sample and this subject was counted with the group carrying two or more active genes. ‡Significances according to the Jonckheere–Terpstra nonparametric trend test; p values above 0.15 are given as NS.NS: Nonsignificant.

Table 1. AUC0-last of R- and S-carvedilol and their metabolites in relation to CYP2D6 genotype.

Substance Median (interquartile ranges [q1–q3] [nmol•h/l]) p-value (trend)‡

Number of active CYP2D6 genes†

0 (n = 13) 0.5 (n = 12) 1 (n = 22) 1.5 (n = 14) ≥2 (n = 38)

R-carvedilol 1356 (955–1552) 558 (384–878) 605 (407–805) 430 (289–1053) 368 (240–455) < 0.001S-carvedilol 279 (213–378) 158 (116–226) 212 (162–306) 218 (142–373) 143 (109–209) 0.001R-desmethylcarvedilol 91.3 (66.3–148.7) 41.1 (20.9–50.6) 42.7 (33.3–74.0) 41.5 (31.8–64.2) 22.6 (12.0–50.6) < 0.001S-desmethylcarvedilol 44.2 (14.3–86.5) 15.3 (10.7–41.3) 41.4 (20.7–88.0) 23.3 (11.1–33.6) 14.6 (9.13–36.0) 0.011R-4’-OH-carvedilol 16.7 (14.2–19.7) 13.9 (6.9–18.6) 17.3 (11.6–25.8) 17.9 (13.6–24.8) 19.2 (16.1–25.3) 0.013S-4’-OH-carvedilol 32.8 (27.6–37.6) 23.9 (11.1–35.3) 30.8 (23.3–50.4) 32.5 (23.3–52.4) 31.5 (22.2–43.7) NSR-5’-OH-carvedilol 32.7 (23.2–39.2) 22.6 (10.8–34.3) 27.5 (19.9–40.2) 24.1 (18.0–36.9) 24.2 (18.7–32.1) NSS-5’-OH-carvedilol 39.1 (30.6–52.5) 26.8 (12.5–45.5) 34.5 (26.6–44.1) 31.3 (17.6–37.0) 25.1 (21.4–39.4) 0.071†For this summarizing analysis, 0.0 activity units were attributed to CYP2D6 alleles *3, *4, *5, *6 and *7, 0.5 activity units were attributed to alleles *9, *10, *17 and *41 and 1.0 activity unit was attributed to *1, *2 and *35. There was only one ultrarapid metabolizer in this sample and this subject was counted with the group carrying two or more active genes. ‡Significances according to the Jonckheere–Terpstra nonparametric trend test; p-values above 0.15 are given as NS.NS: Nonsignificant.



significantly depend on bodyweight (r = 0.37 and 0.33, respectively; p < 0.001). A total of 1 kg less of bodyweight resulted in 0.25 bpm more in maximum resting heart rate reduction.

According to prior knowledge, only S-carvedilol acts via ADRB1 antagonism on heart rate. Constants of the S-carvedilol pharmacoki-netic/pharmacodynamic relationship are given in Supplementary table 4. Neither EC

50 nor E

max was sig-

nificantly associated with ADRB1 or ADRB2 gen-otypes or haplotypes. Surprisingly, EC

50 tended

to be slightly larger in carriers of CYP2D6 low-activity genotypes (0 or 0.5 active alleles), with a decrease by 1.6 nM per active gene according to linear regression analysis. This may be an indi-cation of a contribution of CYP2D6-dependent active metabolites to the carvedilol heart rate reduction, but r2 was only 0.04 (p = 0.03).

Exercise heart rate reduction by carvedilol was the primary pharmacodynamic parameter of our study. Within the tested range up to submaximal workload, heart rate increased lin-early, with a median (range) of 0.47 (0.17–1.76) beats/Watt prior to carvedilol (illustrated also in Supplementary Figure 2). To control for differences in physical training we normalized to the work-load, producing a heart rate of 140 bpm prior to carvedilol (HR

load140). The maximum decrease of

this HRload140

by carvedilol was 20.0 (37.8–2.4) bpm and the median (range) integrated change by carvedilol over 24 h was -13,092 (-34,581 to -11,169) heartbeats. These carvedilol effects increased with increasing number of ADRB1 H1 and H3 haplotypes (p = 0.01 and 0.04 for inte-grated and maximum effect on exercise heart rates, trend-test [Figure 4]).

n Blood pressureThe effects of CYP2D6 and adrenoreceptor geno-types on systolic blood pressure (SBP) reduction by carvedilol were analyzed similarly to the heart rates (descriptive data given in Supplementary

table 4). Carvedilol effects on resting SBP were

not correlated with gender, age, bodyweight or CYP genotypes. However, maximum reduction in resting blood pressure was dependent on the ADRB1 genotype (p = 0.01) with mean reduc-tions of 14.1, 10.8 and 8.1 in carriers of the codon 49 Gly/Gly, Ser/Gly and Ser/Ser genotypes, respectively. There were no baseline differences related to the ADRB1 Arg389Gly genotype or the ADRB2 Arg16Gly or Glu27Gln genotypes. However, reduction in resting blood pressure by carvedilol was significantly greater in carriers of ADRB2 Gln27 compared with Glu27 carriers (maximum reduction, p = 0.001; 24-h reduc-tion, p = 0.003, illustrated in Figure 5). Mean SBP reductions were 6.0, 9.1 and 12.2 mmHg in carriers of Glu27Glu, Glu27Gln and Gln27Gln. The effect of the ADRB2 Gln27 genotype was quantitatively essentially the same after control-ling for gender, age, bodyweight, body height and ADRB1 and CYP genotypes. The ADRB2 Arg16Gly polymorphism alone had no significant effect on SBP, but ADRB2 Arg16Gly–Glu27Gln diplotypes significantly predicted the integrated SBP reduction (p < 0.0005) with a reduction linearly depending on the number of ADRB2 Gly16–Gln27 haplotypes (medians of 16, -64 and -111 mmHg*h in carriers of 0, 1 and 2 Gly16–Gln27 haplotypes, p = 0.0002 for trend).

The maximum and integrated carvedilol effects on exercise-induced systolic blood pres-sure (summarized in Supplementary table 4) were not correlated with any of the studied demographic and genetic factors.

In addition to bicycle exercise stress testing, we measured heart rate and blood pressure by the orthostatic stress test. Volunteers spent at least 5 min in the supine position and their blood pres-sure was then measured. The volunteers next got into the upright position, and their blood pres-sure was measured every minute while standing. The test was administered prior to and 3, 6 and 9 h after carvedilol (Supplementary Figure 1). None of the supine or upright heart rate or blood pressure

Table 3. R- and S-carvedilol immediate release population pharmacokinetic parameters: CYP2D6-independent parameters.

Parameter Unit R-carvedilol S-carvedilol

Mean SE Mean SE

V (central volume of distribution)/73 kg† l 142.8 6.7 285.3 14.31

K12 (rate, central to peripheral compartment) h-1 0.075 0.005 0.115 0.006

K21 (rate, peripheral to central compartment) h-1 0.059 0.006 0.057 0.007

Cl0 (CYP2D6-independent clearance)/73 kg‡ l/h 27.12 1.09 51.85 2.27†Dependent on bodyweight as 1.96*bodyweight [kg] for R-carvedilol and 3.91*bodyweight for S-carvedilol within the range of 50–100 kg. ‡1.086*bodyweight3/4 for R- and 2.076*bodyweight3/4 for S-carvedilol, within the range of 50–100 kg.SE: Standard error.



measurements differed significantly among CYP2D6 genotypes. However, supine heart rate prior to carvedilol and at 6 and 9 h after carvedilol depended on the ADRB1 Ser49Gly polymor-phism (p = 0.02, 0.009 and 0.003); in addition, baseline and 6-h upright heart rates differed significantly between the Ser49Gly genotypes. The supine diastolic blood pressures measured before and 6 and 9 h after carvedilol also differed between the Ser49Gly genotypes (p = 0.02, 0.006 and 0.009), but the individual differences in heart rates and blood pressure data (measurements after carvedilol minus baseline measurements) did not differ between the ADRB1 genotypes. By con-trast, there were no baseline differences related to the ADRB2 genotypes, but upright diastolic blood pressure at 3, 6 and 9 h after carvedilol differed between the ADRB2 Glu27Gln geno-types (p = 0.02, 0.006 and 0.003). Quantitatively, these differences were small, with medians of

57.0 versus 52.4, 52.7 versus 51.4, and 56.4 ver-sus 53.8 mmHg at 3, 6 and 9 h after carvedilol when comparing homozygous Glu27Glu with homozygous Gln27Gln carriers.

n Carvedilol adverse eventsThere were no serious adverse events after admin-istration of carvedilol to the 110 volunteers. In total, 15 volunteers noted dizziness, mostly when standing up during the orthostatic stress test. Dizziness, fatigue and absent-mindedness occurred more often after carvedilol compared with the baseline, and maxima of these adverse events were between 0 and 6 h after carvedilol administration. All other adverse events appeared with similar frequencies or with a lower frequency after carvedilol than at baseline. Dizziness, fatigue and absent-mindedness after carvedilol were not significantly associated with CYP2D6, ADRB1 or ADRB2 genotypes (c2-test).

Table 5. S-carvedilol CYP2D6 allele-specific clearances and genotype-specific clearances.

Allele Allele-specific clearances†

Genotype-specific oral clearances, Cl/F (SE) [l/h]

Cl *1 *2 *3–*7 *9 *10 *17 *35 *41

*1 10.9 (1.9) 217 (21) 202 (21) 158 (17) 174 (19) 174 (19) 174 (19) 183 (21) 174 (19)

*2 8.33 (2.8) 188 (21) 145 (17) 161 (19) 161 (19) 161 (19) 169 (21) 161 (19)

*3–*7 0.00 108 (14) 121 (15) 121 (15) 121 (15) 129 (17) 121 (15)

*9 3.23 (2.0)‡ 136 (17) 136 (17) 136 (17) 144 (17) 136 (17)

*10 3.23 (2.0) 136 (17) 136 (17) 144 (17) 136 (17)

*17 3.23 (2.0) 136 (18) 144 (17) 136 (17)

*35 4.85 (3.4) 152 (18) 144 (18)

*41 3.23 (2.0) 136 (17)†In contrast to the genotype-specific oral clearances, these allele-specific clearances per one allele do not include the relative contribution of CYP2D6 to the first-pass metabolism (the genotype-specific relative bioavailabilities are given in Supplementary table 3).‡S-carvedilol clearances for CYP2D6 alleles *9, *10, *17 and *41 were indistinguishable.SE: Standard error.

Table 4. R-carvedilol CYP2D6 allele-specific clearances and genotype-specific clearances.

Allele Allele-specific clearances†

Genotype-specific oral clearances, Cl/F (SE) [l/h]

Cl (SE) *1 *2 *3–*7 *9 *10 *17 *35 *41

*1 12.9 (1.2) 112 (7.7) 93.5 (7.3) 64.1 (5.5) 87.6 (7.9) 88.0 (7.9) 68.5 (6.2) 84.3 (7.3) 77.7 (6.7)

*2 8.31 (1.6) 76.6 (6.9) 50.2 (5.1) 71.3 (7.3) 71.6 (7.3) 54.1 (5.8) 68.3 (6.8) 62.4 (6.3)

*3–*7 0.0 29.4 (3.5) 46.0 (5.5) 46.2 (5.5) 32.4 (4.1) 43.5 (5.0) 38.9 (4.5)

*9 6.78 (2.4) 66.2 (7.7) 66.5 (7.7) 49.7 (6.1) 63.3 (7.2) 57.6 (6.6)

*10 6.87 (2.4) 66.8 (6.1) 49.9 (6.1) 63.6 (7.2) 57.9 (6.6)

*17 1.34 (1.6) 35.5 (4.6) 47.2 (5.6) 42.3 (5.1)

*35 5.87 (2.0) 60.4 (6.7) 54.9 (6.1)

*41 4.05 (1.7) 49.6 (5.7)†In contrast to the genotype-specific oral clearances, these allele-specific clearances per one allele do not include the relative contribution of CYP2D6 to the first-pass metabolism (the genotype-specific relative bioavailabilities are given in Supplementary table 3).SE: Standard error.



DiscussionThe present study was initiated to clarify whether and how precisely the individual carvedilol doses might be adjusted based on CYP2D6 and b adrenoreceptor genotypes. There are a number of pharmacokinetic studies that reproducibly show some impact of CYP2D6 polymorphisms on carvedilol pharmacokinetics [3–6], but there are only scarce combined analyses of carvedilol PKPDPG. Compared with earlier studies our study had a larger sample size. The known effect of CYP2D6 on carvedilol pharma-cokinetics was confirmed, but the CYP2D6 polymorphisms had no apparent effect on heart rate, blood pressure or side effects. The

detailed pharmacokinetics, pharmacodynamics analysis provided several possible explanations for why the CYP2D6-related pharmacokinetic variation had no impact on pharmacodynam-ics: First, the CYP2D6 effect, particularly on the S-enantiomer, was small (FigureS 1 & 2); only approximately 15% of pharmacokinetic vari-ation was explained by CYP2D6. Second, the active metabolite R-4 -́OH carvedilol (table 2) may contribute to carvedilol clinical effects. Thus, the higher activity in poor metaboliz-ers due to higher parent substance concentra-tions may be compensated by the lower activity in poor metabolizers due to lower concentra-tions of the active metabolite. This is not pure

Plasma S-carvedilol concentration (nmol/l)

Hea

rt r

ate

(bp

m)

Hea

rt r

ate

(bp

m)

Time after administration of 25 mg carvedilol (h)

130

110

90

70

500 10 20 30 40

130

110

90

70

500 12 24 36 48

Exercise induced Exercise induced

Resting Resting

A B

Figure 3. Interindividual variation in the carvedilol concentration–effect relationship. (A) Shows individual (in black) and population mean predictions (in purple) of heart rate in relation to carvedilol plasma concentrations. The three purple lines refer to the means of carriers of the ADRB1 Ser49Gly genotypes (see Figure 4 for more details). The upper array of curves in this figure shows the heart rates normalized to a workload, which resulted in a heart rate of 140 bpm without carvedilol. Thus, all exercise heart rates were 140 at zero carvedilol plasma concentration but the individual reactivity to carvedilol showed great interindividual variation. Only a minor fraction of this variation was explained by the ADRB1 genotype (purple lines). The lower array of curves shows the interindividual variation in resting heart rates. The relatively flat concentration–effect relationship at higher carvedilol concentrations may be one reason why CYP2D6 genotype was not reflected in carvedilol effects or side effects. While (A) illustrates the pharmacodynamics variation only, (B) illustrates the combined result of pharmacokinetic and pharmacodynamic variation. Population mean predictions for the CYP2D6 poor metabolizer genotype and the three ARDB1 Ser49Gly genotypes are depicted in blue, and the corresponding mean predictions for the CYP2D6 extensive metabolizer genotype and the three ADRB1 genotypes are depicted in red. These predictions, based on the S-carvedilol pharmacokinetics and concentration–effect relationship, indicate that the CYP2D6-related differences in carvedilol pharmacodynamics may be only small.



speculation given that 4 -́OH-carvedilol may be approximately 13-times more active than the parent drug [22] and a recent study showed even higher adjusted carvedilol doses in poor metabo-lizers [9], which might indicate that CYP2D6 partially plays a bioactivating role in carvedilol therapy. Third, a flat concentration–effect rela-tionship may explain the lack of repercussion of CYP2D6 variation in carvedilol effects. As illustrated by the concentration–effect curves (Figure 3), at S-carvedilol concentrations above 10 nmol/l, a difference as big as 100% in plasma concentration may only result in a resting heart rate difference of approximately 2 bpm. Finally, there appears to be great interindividual vari-ation in the affinity and maximum carvedilol effects (more than tenfold variation in EC

50 and

Emax

values [Figure 3 & Supplementary table 4]). Irrespective of the apparent lack of CYP2D6

impact on carvedilol therapy, there are interest-ing results regarding the CYP2D6 genotype–phenotype relationship. The common CYP2D6 variants *2 and *35 had a significantly lower carvedilol metabolizing activity than CYP2D6*1 (Figure 2 & tableS 4 & 5). Both alleles are usually considered as normally active, and with debriso-quine as the substrate, the allele CYP2D6*35 was even associated with ultra-rapid metabolism [23]. Both CYP2D6*2 and *35 differ from the wild-type enzyme in Arg296Cys and Ser486Thr (see Supplementary table 1 for an overview of the molecular differences between the CYP2D6 alleles). The approximately 30% lower activity of CYP2D6*2 compared with CYP2D6*1 found with carvedilol is compatible with earlier data on dextromethorphan demethylation [24–26]. Such possibly substrate-specific differences among the CYP2D6 alleles *1, *2, *9, *10, *17 and *35 may be the background of a long-known dissociation between debrisoquine and metoprolol hydroxy-lation phenotypes noted earlier in Africans [27]. However, from the data of our present study we cannot differentiate whether the differ-ences between CYP2D6*1, *2 and *35 are due to different catalytic activities, lower protein or mRNA expression rates or stabilities.

Simplified and easily comprehensible classifi-cation schemes for the multiple CYP2D6 geno-types have been described earlier [28–31] and are important for clinical application of CYP2D6 genotyping. We also used this principle for plan-ning of the present study and for the analyses presented in tableS 1 & 2. However, as presented in tableS 3, 4 & 5, a more detailed classification may be more appropriate, and only activities conferred by the alleles CYP2D6*9 and *10 were really

intermediate between the active allele *1 and the deficient allele *4, while *17 and *41 were closer to zero.

A third notable finding was the unexplained wide pharmacokinetic variation within spe-cific CYP2D6 genotype groups (FigureS 1 & 2). At maximum, 35.7% of the variation in total clearance of R-carvedilol and 15.3% of the variation in total clearance of S-carvedilol was explained by CYP2D6 genotype. The genetic variation in CYP2C9 and CYP2C19, though possibly relevant according to in vitro data [2], did not correlate with any pharmacokinetics or pharmacodynamics data in our study, which confirms data described earlier [32]. Thus, we sought an explanation for the remaining vari-ability. Carvedilol clearances and volumes of distribution depended on bodyweight, and even heart rate and blood pressure reduction was significantly dependent on bodyweight. Some variation might be explained by poly-morphisms in genes coding for the glucuron-osyltransferases UGT1A1, UGT2B4 and UGT2B7 [33] or the ABC drug transporters MDR1 and MRP2 [6], but population frequen-cies of strongly functional polymorphisms in these genes are relatively low. Because of that and because of the multiplicity of UGT1A1, -2B4, -2B7 and MDR1 and MRP2 genotypes possibly involved, we did not include these genotypes in our analysis.

Exercise stress testing with ECG recording is a standard approach to measure the effects of b-blockers in human clinical pharmacology. In this analysis we did not want to limit exer-cise performance in the repeated exercise stress tests by increasing muscular fatigue, and thus we shortened the times at specific workloads to 1 min. With this approach, steady state in heart rate or blood pressure is not achieved at a spe-cific workload, but integrated data analysis using an approach similar to that described earlier [34] allowed us to predict heart rates over the entire range of workloads tested. Supplementary Figure 2 illustrates how well this parsimonious model predicted the heart rates during exercise stress testing with three parameters only (HF0, HFM and KE) and thus summarized dozens of heart rate and blood pressure measurements. The extensive pharmacokinetics, pharmacodynam-ics monitoring of our volunteers may go beyond what was performed in several earlier studies and, therefore, it was interesting to see how the parameters describing the pharmacokinetics, pharmacodynamics relationship correlated with b

1 and b

2 adrenoreceptor genotypes.



Two frequent amino acid polymorphisms in the b

1 adrenoreceptor, Ser49Gly and Arg389Gly,

had a significant effect on baseline resting heart rate independent from carvedilol. As illustrated in Figure 4, the effect of Ser49Gly was stronger than the effect of Arg389Gly. Of course it can-not be ruled out that these differences were due to random variation, but our study sample was specifically enriched for carriers of rare ADRB1 genotypes, and the apparent gene-dose related effect of the ADRB1 variants (Figure 4) also argues against pure chance. Higher heart rates in Gly49 carriers have also been found earlier in African

and Caucasian samples [35], but inverse effects in Asian populations have been described [36], and more recent genome-wide association studies apparently did not find major effects of ADRB1 on resting heart rate [37]. The latter negative finding from a genome-wide association study should not detract from further clinical func-tional analyses of the impact of the ADRB1 genotypes. Apparently the functional conse-quences of the ADRB1 genotypes depend on environmental factors or epistatic effects, which differed between the studies and which are not yet fully understood.

110

100

90

80

70

60

Gly49Gly

Gly49Ser

Ser49Ser

Gly49Gly

Gly49Ser

Ser49Ser

Gly49Gly

Gly49Ser

Ser49Ser

Gly389Gly Gly389Arg Arg389Arg

Res

tin

g h

eart

rat

e (b

eats

/min

)

n = 8 12 31 4 19 35

0

-10

-20

-30

-40

Gly49Gly

Gly49Ser

Ser49Ser

Gly49Gly

Gly49Ser

Ser49Ser

Gly49Gly

Gly49Ser

Ser49Ser


Inte

gra

ted

exe

rcis

e h

eart

rat

e re

du

ctio

n (

bea

ts/2

4 h

)

Max

. exe

rcis

e h

eart

rat

ere

du

ctio

n (

bea

ts/m

in)

10,000

0

-10,000

-20,000

-30,000

-40,000Gly49Gly

Gly49Ser

Ser49Ser

Gly49Gly

Gly49Ser

Ser49Ser

Gly49Gly

Gly49Ser

Ser49Ser


n = 8 12 31 4 19 35

n = 8 12 31 4 19 35

A B

C

Figure 4. Heart rate in relation to ADRB1 genotype. (A) Dependence of baseline resting heart rates from ADRB1 polymorphisms Ser49Gly and Arg389Gly. In two-way analysis of variance, both ADRB1 polymorphisms (Ser49Gly and Arg389Gly) were significantly associated with resting heart rate (p = 0.001 and p = 0.04 for Ser49Gly and Arg389Gly, respectively). The combinations of the codon 49 and codon 389 polymorphisms are referred to as haplotypes H1 (Ser49Arg389), H2 (Ser49Gly389) and H3 (Gly49Arg389). (B) Combined effects of the two ADRB1 genotypes Ser49Gly and Arg389Gly on maximum exercise heart rate reduction by carvedilol. (C) Combined effects of the two ADRB1 genotypes Ser49Gly and Arg389Gly on 24-h integrated exercise heart rate reduction by carvedilol. Note that maximum and integrated heart rate reductions were inversely correlated with baseline heart rates.



Earlier studies have found that the cardiac b

2 adrenoreceptor blockade may contribute to

carvedilol effects [19]. The ADRB2 Glu27 vari-ant appears to be resistant to agonist-promoted downregulation [20] and such Glu27 carriers showed a better response to carvedilol than Gln27 carriers [21], a finding that was, however, not confirmed in a larger study including signifi-cantly more patients [14]. In our healthy volunteer sample, there were no baseline differences related

to ADRB2 genotypes, but there was a greater blood pressure reduction in carriers of the Gln27 genotype. This effect was quantitatively small, but was seen consistently in the resting SBP meas-urements performed on separate occasions prior to the exercise stress tests (Figure 5) and as part of the orthostatic stress tests(Supplementary table 4).

In this study we intentionally restricted the analysis to three genes and the most frequently studied variants in these genes because we wanted to clarify whether these CYP2D6, ADRB1 and ADRB2 genotypes might be valid biomark-ers in carvedilol-related research and therapy. As illustrated in figure 1 by Evans and Relling, in the path of translating functional genomics into rational therapeutics [38] it is the PKPDPG analysis that has the full power to elucidate the medical role of genomic variation. Analyzing only pharmaco kinetics in relation to gene variants ends with surrogate parameters with equivocal clini-cal significance. On the other hand, analyzing pharmacodynamics or clinical outcomes without knowing the pharmacokinetics, does not really allow differentiating and understanding how and how much genes modulate drug effects. This was the background behind the design (Supplementary

Figure 1) of our PKPDPG study presented here.Both studies in healthy volunteers and studies

in patients have their own advantages and limita-tions. An advantage of healthy volunteer studies is that we do not have to deal with the great variation usually seen in patients with diseases and such volunteer studies should thus be help-ful in elucidating the genotype–phenotype rela-tionships for many polymorphisms in humans. But the medical impact of polymorphisms may change depending on age, disease states, medi-cations and gene–gene and gene–environment interactions, which are still not yet fully under-stood. This may be the reason why the medical role of ADRB1 and ADRB2 polymorphisms in clinical medicine is still controversial.

ConclusionAs a conclusion from our carvedilol pharmacoki-netics, pharmacodynamics analysis, it is unlikely that CYP2D6 plays a relevant role in clinical ther-apy with carvedilol. However, concerning other drugs, the substrate-dependent differential activity of CYP2D6 alleles *1, *2, *9, *10, *17, *35 and *41 is notable. The adreno receptor ADRB1 Ser49Gly and ADRB2 Glu27Gln genotypes modulated the cardiovascular effects of carvedilol. However, there is no consistent directionality of the genotype–phenotype relationships of ADRB1 and ADRB2 polymorphisms. Thus, at present an application

0

-5

-10

-15

-20

-25

-30

Max

. res

tin

g S

BP

red

uct

ion

(m

mH

g)

Glu27Glu Glu27Gln Gln27Gln

Glu27Glu Glu27Gln Gln27Gln

100

0

-100

-200

-300

-400

Inte

gra

ted

res

tin

g S

BP

red

uct

ion

(m

mH

g•2

4 h

)

n = 19 51 37

n = 19 51 37

A

B

Figure 5. Systolic blood pressure and adrenergic b2-receptor polymorphisms. (A) Maximum reduction of resting blood pressure dependent on the b2-receptor Glu27Gln polymorphism. (B) Effect of the ADRB2 Glu27Gln polymorphism on integrated 24-h blood pressure reduction by carvedilol. Notably there were not baseline differences in blood pressure between the studied ADRB2 genotypes. The trend of increasing blood pressure reduction with increasing number of Gln27 alleles was particularly clear in the subgroups who were carriers of the Gly allele in ADRB2 codon 16. SBP: Systolic blood pressure.



of ADRB genotypes in individualized medicine is not yet possible. Further studies are needed to understand the medical impact of these polymor-phisms, an impact that may differ depending on age, type and duration of cardiovascular diseases, gene–gene and gene–environment interactions and the specific medications administered.

AcknowledgementsThe authors thank Hannelore Steinmetz and Cornelia Willnow for their excellent technical assistance in concen-tration measurements and we thank Adda Boekhoff, Irmani Beatrice Ismadi, Julia-Patricia Kaup and Kathrin Trang for their important contributions to this study. Many thanks also to Valerie O’Brien for carefully correcting the English language of the manuscript. Roche Germany has kindly provided the carvedilol reference substances.

Financial & competing interests disclosureRoche Molecular Diagnostics has financially supported a part of this study. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No financial writing assistance was utilized in the production of this manuscript.

Ethical conduct of research The authors state that they have obtained appropriate insti-tutional review board approval or have followed the princi ples outlined in the Declaration of Helsinki for all human or animal experimental investigations. In addition, for investi-gations involving human subjects, informed consent has been obtained from the participants involved.

Executive summary

� Predicting the individual chance to respond to or predicting the individual risk to suffer from adverse events from carvedilol is of great medical interest in the treatment of hypertension and heart failure. A combined pharmacokinetic/pharmacodynamic/pharmacogenetic (PKPDPG) analysis is a promising approach to a better understanding of the medical impact of single and multiple genetic polymorphisms.

� In the present PKPDPG analysis, the impact of polymorphisms in CYP2D6 (alleles *3, *4, *5, *6, *9, *10, *17, *35 and *41) and in the b-adrenergic receptor genes ADRB1 (Gly49Ser, Arg389Gly) and ADRB2 (Arg16Gly, Glu27Gln) on carvedilol was analyzed.

Materials & methods � PKPDPG of a 25 mg oral dose of racemic carvedilol were studied in 110 healthy male and female volunteers. � Using prestudy genotyping, the sample was enriched for deficient and intermediate CYP2D6 alleles (*3, *4, *5, *6, *9, *10, *17 and *41)

and the ADRB1 Gly49 and Gly389 alleles, which increased the statistical power of the study concerning functional analysis of these genotypes. � Polymorphisms in CYP2D6 (alleles *1, *2, *3, *4, *5, *6, *9, *10, *17, *35 and *41) and polymorphisms in ADRB1 and ADRB2

were analyzed by PCR and plasma concentrations of carvedilol, desmethylcarvedilol, 4’-OH-carvedilol and 5’-OH-carvedilol were enantioselectively analyzed by high-performance liquid chromatography.

� Carvedilol effects on heart rate and blood pressure were electronically recorded and adverse events were monitored. More than 500 heart rate records and 50 blood pressure records per individual study participant were analyzed in relation to pharmacokinetics and pharmacogenetics by population PKPDPG modeling.

Results � Maximally 36% of the variation in total clearance of R-carvedilol and 15% of the variation in total oral clearance of S-carvedilol was

explained by CYP2D6 genotype. � Differential clearances were identified for the CYP2D6 alleles *1, *2, *9, *10, *17, *35 and *41 with allele-specific systemic R-carvedilol

clearances of 12.9, 8.3, 6.8, 6.9, 1.3, 5.9, and 4.1 l/h. The corresponding systemic allele-specific S-carvedilol clearances were 10.9, 8.3, 3.2, 3.2, 3.2, 4.9 and 3.2 l/h, respectively.

� Compared with poor metabolizers of CYP2D6 substrates, extensive metabolizers had lower concentrations of carvedilol and desmethylcarvedilol but higher concentrations of the active metabolite 4-OH-carvedilol. In extensive metabolizers of CYP2D6 substrates defined by genotype, 74% of R-carvedilol and 50% of S-carvedilol was eliminated via CYP2D6.

� Resting and exercise-induced heart rate, blood pressure and adverse effects before and after administration of carvedilol were not significantly different between the CYP2D6 genotypes.

� The ADRB1 Gly49 allele was associated with higher baseline heart rates. Carvedilol-induced reduction of exercise heart rates was greater in Gly49 and in Arg389 carriers compared with the more frequent Ser49 and Gly389 alleles.

� While baseline heart rate and blood pressure were independent from ADRB2 genotypes, carvedilol-induced reduction of resting blood pressure was greater in carriers of the ADRB2 Gln27 allele compared with the Glu27 allele.

Conclusion � In drugs metabolized by CYP2D6, the medical impact of the CYP2D6 polymorphism is of concern, but may not be relevant in some

instances, as shown here for carvedilol. � Not only, does the difference between deficient and active CYP2D6 alleles have to be taken into consideration, but also the potentially

differential activities of CYP2D6 alleles *1, *2, *9, *10, *17, *35 and *41. � The Ser49Gly and Arg389Gly polymorphisms in the adrenergic b1 receptor and the Gln27Glu polymorphism in the adrenergic b2

receptor modulated carvedilol effects on heart rate and blood pressure.



BibliographyPapers of special note have been highlighted as:n of interest

1 Bartsch W, Sponer G, Strein K et al.: Pharmacological characteristics of the stereoisomers of carvedilol. Eur. J. Clin. Pharmacol. 38(Suppl. 2), S104–S107 (1990).

2 Oldham HG, Clarke SE: In vitro identification of the human cytochrome P450 enzymes involved in the metabolism of R(+)- and S(-)-carvedilol. Drug Metab. Dispos. 25(8), 970–977 (1997).

n Comprehensive in vitro identification of the cytochrome P450 isoenzymes potentially relevant for carvedilol.

3 Zhou HH, Wood AJ: Stereoselective disposition of carvedilol is determined by CYP2D6. Clin. Pharmacol. Ther. 57(5), 518–524 (1995).

n First clinical study demonstrating the impact of CYP2D6 for carvedilol pharmacokinetics.

4 Honda M, Nozawa T, Igarashi N et al.: Effect of CYP2D6*10 on the pharmacokinetics of R- and S-carvedilol in healthy Japanese volunteers. Biol. Pharm. Bull. 28(8), 1476–1479 (2005).

5 Takekuma Y, Takenaka T, Kiyokawa M et al.: Contribution of polymorphisms in UDP-glucuronosyltransferase and CYP2D6 to the individual variation in disposition of carvedilol. J. Pharm. Pharm. Sci. 9(1), 101–112 (2006).

6 Giessmann T, Modess C, Hecker U et al.: CYP2D6 genotype and induction of intestinal drug transporters by rifampin predict presystemic clearance of carvedilol in healthy subjects. Clin. Pharmacol. Ther. 75(3), 213–222 (2004).

n Clinical study clearly showing the impact of efflux transporters on carvedilol pharmacokinetics.

7 Saito M, Kawana J, Ohno T et al.: Population pharmacokinetics of R- and S-carvedilol in Japanese patients with chronic heart failure. Biol. Pharm. Bull. 33(8), 1378–1384 (2010).

8 Graff DW, Williamson KM, Pieper JA et al.: Effect of fluoxetine on carvedilol pharmacokinetics, CYP2D6 activity, and autonomic balance in heart failure patients. J. Clin. Pharmacol. 41(1), 97–106 (2001).

9 Baudhuin LM, Miller WL, Train L et al.: Relation of ADRB1, CYP2D6, and UGT1A1 polymorphisms with dose of, and response to, carvedilol or metoprolol therapy in patients with chronic heart failure. Am. J. Cardiol. 106(3), 402–408 (2010).

10 Dorn GW 2nd, Liggett SB: Mechanisms of pharmacogenomic effects of genetic variation within the cardiac adrenergic network in heart failure. Mol. Pharmacol. 76(3), 466–480 (2009).

n Recent review on pharmacogenetics of adrenergic receptor and signaling pathways in relation to heart failure.

11 Liggett SB, Mialet-Perez J, Thaneemit-Chen S et al.: A polymorphism within a conserved b(1)-adrenergic receptor motif alters cardiac function and b-blocker response in human heart failure. Proc. Natl Acad. Sci. USA 103(30), 11288–11293 (2006).

12 Mason DA, Moore JD, Green SA et al.: A gain-of-function polymorphism in a G-protein coupling domain of the human b

1-adrenergic receptor. J. Biol. Chem. 274(18),

12670–12674 (1999).

13 Rathz DA, Brown KM, Kramer LA et al.: Amino acid 49 polymorphisms of the human b

1-adrenergic receptor affect agonist-

promoted trafficking. J. Cardiovasc. Pharmacol. 39(2), 155–160 (2002).

14 Sehnert AJ, Daniels SE, Elashoff M et al.: Lack of association between adrenergic receptor genotypes and survival in heart failure patients treated with carvedilol or metoprolol. J. Am. Coll. Cardiol. 52(8), 644–651 (2008).

n Clinical studies showing that adrenergic b1,

b2, and a

2c polymorphisms may not have a

relevant impact on the clinical outcome of heart failure therapy with carvedilol or metoprolol.

15 Small KM, Wagoner LE, Levin AM et al.: Synergistic polymorphisms of b

1- and

a2C

-adrenergic receptors and the risk of congestive heart failure. N. Engl. J. Med. 347(15), 1135–1142 (2002).

16 Rochais F, Vilardaga JP, Nikolaev VO et al.: Real-time optical recording of b

1-adrenergic

receptor activation reveals supersensitivity of the Arg389 variant to carvedilol. J. Clin. Invest. 117(1), 229–235 (2007).

17 Johnson JA, Zineh I, Puckett BJ et al.: b 1-adrenergic receptor polymorphisms and antihypertensive response to metoprolol. Clin. Pharmacol. Ther. 74(1), 44–52 (2003).

18 Pacanowski MA, Gong Y, Cooper-Dehoff RM et al.: b-adrenergic receptor gene polymorphisms and b-blocker treatment outcomes in hypertension. Clin. Pharmacol. Ther. 84(6), 715–721 (2008).

n Clinical study showing a higher mortality in hypertensive carriers of the adrenergic b

1

receptor haplotype Ser49Arg389.

19 Molenaar P, Christ T, Ravens U et al.: Carvedilol blocks b

2- more than b

1-

adrenoceptors in human heart. Cardiovasc. Res. 69(1), 128–139 (2006).

20 Green SA, Turki J, Innis M et al.: Amino-terminal polymorphisms of the human b

2-adrenergic receptor impart distinct agonist-

promoted regulatory properties. Biochemistry 33(32), 9414–9419 (1994).

n One of the first publications showing functional consequences of the in the adrenergic b

2 receptor Glu27Gln

polymorphism.

21 Kaye DM, Smirk B, Williams C et al.: b-adrenoceptor genotype influences the response to carvedilol in patients with congestive heart failure. Pharmacogenetics 13(7), 379–382 (2003).

22 Dunn CJ, Lea AP, Wagstaff AJ: Carvedilol. A reappraisal of its pharmacological properties and therapeutic use in cardiovascular disorders. Drugs 54(1), 161–185 (1997).

23 Lovlie R, Daly AK, Matre GE et al.: Polymorphisms in CYP2D6 duplication-negative individuals with the ultrarapid metabolizer phenotype: a role for the CYP2D6*35 allele in ultrarapid metabolism? Pharmacogenetics 11(1), 45–55 (2001).

24 McElroy S, Sachse C, Brockmöller J et al.: CYP2D6 genotyping as an alternative to phenotyping for determination of metabolic status in a clinical trial setting. AAPS PharmSci. 2(4), E33 (2000).

25 Abduljalil K, Frank D, Gaedigk A et al.: Assessment of activity levels for CYP2D6*1, CYP2D6*2, and CYP2D6*41 genes by population pharmacokinetics of dextromethorphan. Clin. Pharmacol. Ther. 88(5), 643–651 (2010).

n Study clearly showing functional differences between the two frequent CYP2D6 alleles CYP2D6*1 and CYP2D6*2.

26 Bapiro TE, Hasler JA, Ridderstrom M et al.: The molecular and enzyme kinetic basis for the diminished activity of the cytochrome P450 2D6.17 (CYP2D6.17) variant. Potential implications for CYP2D6 phenotyping studies and the clinical use of CYP2D6 substrate drugs in some African populations. Biochem. Pharmacol. 64(9), 1387–1398 (2002).

n Biochemical analysis of substrate-specific catalytic differences of the CYP2D6*17 variant very frequent in Africans.

27 Lennard MS, Iyun AO, Jackson PR et al.: Evidence for a dissociation in the control of sparteine, debrisoquine and metoprolol metabolism in Nigerians. Pharmacogenetics 2(2), 89–92 (1992).

n Presents early evidence for substrate-specific interethnic differences in CYP2D6 activity, which may now be explained by substrate specific differences in the activity of specific CYP2D6 alleles.



28 Steimer W, Zopf K, von Amelunxen S et al.: Allele-specific change of concentration and functional gene dose for the prediction of steady-state serum concentrations of amitriptyline and nortriptyline in CYP2C19 and CYP2D6 extensive and intermediate metabolizers. Clin. Chem. 50(9), 1623–1633 (2004).

n Introduces the term quantitative functional gene dose, which helps to summarize the numerous CYP2D6 allele variants into one informative number per individual.

29 Kirchheiner J, Heesch C, Bauer S et al.: Impact of the ultrarapid metabolizer genotype of cytochrome P450 2D6 on metoprolol pharmacokinetics and pharmacodynamics. Clin. Pharmacol. Ther. 76(4), 302–312 (2004).

n Population pharmacokinetic–pharmacodynamic analysis of the b-blocker metoprolol with emphasis on ultra-rapid metabolizers.

30 Kirchheiner J, Fuhr U, Brockmöller J: Pharmacogenetics-based therapeutic recommendations – ready for clinical practice? Nat. Rev. Drug Discov. 4(8), 639–647 (2005).

n Perspectives paper summarizing the need and possible ways for implementation of pharmacogenetic diagnostics in medical practice.

31 Gaedigk A, Simon SD, Pearce RE et al.: The CYP2D6 activity score: translating genotype information into a qualitative measure of phenotype. Clin. Pharmacol. Ther. 83(2), 234–242 (2008).

32 Honda M, Ogura Y, Toyoda W et al.: Multiple regression analysis of pharmacogenetic variability of carvedilol disposition in 54 healthy Japanese volunteers. Biol. Pharm. Bull. 29(4), 772–778 (2006).

33 Ohno A, Saito Y, Hanioka N et al.: Involvement of human hepatic UGT1A1, UGT2B4, and UGT2B7 in the glucuronidation of carvedilol. Drug Metab. Dispos. 32(2), 235–239 (2004).

34 Riley MS, Porszasz J, Engelen MP et al.: Responses to constant work rate bicycle ergometry exercise in primary pulmonary hypertension: the effect of inhaled nitric oxide. J. Am. Coll. Cardiol. 36(2), 547–556 (2000).

35 Wilk JB, Myers RH, Pankow JS et al.: Adrenergic receptor polymorphisms associated with resting heart rate: the HyperGEN Study. Ann. Hum. Genet. 70(Pt 5), 566–573 (2006).

36 Ranade K, Jorgenson E, Sheu WH et al.: A polymorphism in the b1 adrenergic receptor is associated with resting heart rate. Am. J. Hum. Genet. 70(4), 935–942 (2002).

37 Eijgelsheim M, Newton-Cheh C, Sotoodehnia N et al.: Genome-wide association analysis identifies multiple loci related with resting heart rate. Hum. Mol. Genet. 19(19), 3885–3894 (2010).

38 Evans WE, Relling MV: Pharmacogenomics: translating functional genomics into rational therapeutics. Science 286(5439), 487–491 (1999).

n Inspiring review on pharmacogenomics, cited here because of its figure 1, which summarizes why both pharmacokinetics and pharmacodynamics (and possibly clinical outcomes) should be studied.

n Website101 CYP2D6 allele nomenclature

www.cypalleles.ki.se/cyp2d6.htm

carvedilol pharmacokinetics and pharmacodynamics in relation to

Documents