Escolar Documentos
Profissional Documentos
Cultura Documentos
1
AR, could also alter clinical responses to -blocker
therapy. There are 2 common nonsynonymous poly-
morphisms in the coding region of the ADRB1 gene,
Gly49Ser, which is located in the extracellular termi-
nus, an area suggested to be important for receptor
internalization and trafcking, and Gly389Arg, which
is located in the intracellular carboxy terminus in the
putative G-protein binding site.
21,22
The prevalence of
the Gly49 allele is approximately 15%, with no differ-
ence between white subjects and black subjects,
whereas the Gly389 allele is more frequent in white
subjects (42%) than in black subjects (27%).
22
In vitro studies suggest that both polymorphisms are
functionally important. The Gly49 variant showed de-
creased receptor expression compared with Ser49 after
exposure to isoproterenol (INN, isoprenaline) for
24 hours,
23
and in the Arg389 variant, adenylyl cyclase
activity in response to agonist was 3 times greater than
that of the Gly389 variant.
24
Three studies have shown
that these variants affect responses to -blockers in
vivo.
12-14
In normotensive subjects the resting blood pressure
response to atenolol, a selective
1
AR antagonist, was
signicantly greater in Arg389 homozygous subjects
compared with Gly389 homozygous subjects.
12
A pro-
spective study of 40 hypertensive patients treated with
metoprolol conrmed this observation, showing a 2- to
3-fold greater reduction in diastolic blood pressure in
Arg389 homozygotes compared with Gly389 heterozy-
gotes.
13
However, in addition to the contribution of
variation at position 389, the effect of ADRB1 diplotype
was a signicant predictor of blood pressure response
to metoprolol. Patients with Ser49/Ser49 Arg389/
Arg389 had a decline of 14.7 mm Hg in diastolic blood
pressure as compared with 0.5 mm Hg in patients with
the Gly49/Ser49 Arg389/Gly389 diplotype.
13
In a third
study, there was a larger reduction in resting and exer-
cise heart rates and systolic blood pressure after meto-
prolol in healthy subjects homozygous for Arg389 than
in Gly389 homozygotes.
14
However, despite these
studies in healthy subjects and in hypertensive patients,
there is little information about the effect of ADRB1
genotype and response to -blockers in patients with
heart failure.
CHF results in increased sympathetic stimulation
and decreased -AR responsiveness
25
that is related to
decreased -AR density and reduced maximal
isoproterenol-stimulated adenylate cyclase stimula-
tion.
26
Treatment of patients with heart failure with a
1
AR antagonist was associated with an increase in
myocardial -AR density and improved hemodynamic
responses.
27
Thus variations in the ADRB1 gene that
alter
1
AR expression (the Ser49Gly polymorphism
23
)
or response to agonist (the Arg389Gly polymor-
phism
24
) could potentially be clinically relevant.
Studies in transgenic mice with cardiac overexpres-
sion of the ADRB1 variants found that hemodynamic
responses to -blockade were greater in mice overex-
pressing Arg389.
28
Similarly, in humans with heart
failure, preliminary evidence indicated that the pres-
ence of the Arg389 allele was associated with improve-
ment of left ventricular function during carvedilol treat-
ment.
28
There is, however, little information regarding
the effects of CYP2D6 and ADRB1 genotype on re-
sponse to -blocker therapy in patients with heart fail-
ure.
In this issue of the Journal, Terra et al
29
describe the
effect of variants in 2 candidate genes, CYP2D6 and
ADRB1, on response to -blocker therapy in 61 patients
with CHF. The primary end point was the inability to
tolerate -blocker therapy, dened as the need to dis-
continue metoprolol or the inability to reach the target
dose of 200 mg/d by the end of the titration period. A
secondary outcome was decompensated heart failure, a
composite measure that included any of the following
end points: death, cardiac transplantation, hospitaliza-
tion for worsening heart failure, increase in other heart
failure medications, or the need to discontinue meto-
prolol.
Although genotype did not affect the frequency of
the primary composite outcome measure, a signi-
cantly greater percentage of Gly389 carriers required an
increase in heart failure medications (48% compared
with 14%). Diplotype analysis of polymorphism at po-
sitions 49 and 389 revealed that 52% of patients car-
rying the combination of Ser49Ser/Arg389Gly, 42% of
CLINICAL PHARMACOLOGY & THERAPEUTICS
124 Muszkat and Stein MARCH 2005
Ser49Gly/Arg389Gly patients, 23% of Ser49Ser/
Arg389Arg patients, and no Ser49Gly/Arg389Arg pa-
tients required an increase in concomitant medications
for heart failure. Interestingly, the 2 diplotypes that
included the Gly389 allele and were associated with an
increased need for additional heart failure treatment
were previously found to be associated with the small-
est reduction in diastolic blood pressure in response to
a -blocker.
13
Concordant with the notion that the
Gly389 allele is associated with decreased response to
-blocker was a trend toward a greater 6-minute walk
distance in Arg389 homozygous patients (357 m versus
293 m, P .06). However, ADRB1 variants did not
affect the nal doses achieved over the titration period.
As expected, metoprolol concentrations were signif-
icantly affected by CYP2D6 genotype. Concentrations
of S-metoprolol were 2 to 4 times higher in CYP2D6
PMs compared with EMs; however, as was observed in
patients with hypertension,
15
this pharmacokinetic dif-
ference was not associated with clinically detectable
differences in effect. The lack of a clinically detectable
effect of genetic variations that have marked effects on
metoprolol concentrations is counterintuitive and may
be related to the relatively small number of patients
studied, clinical heterogeneity, or a weak relationship
between plasma concentration and clinical response.
The ndings presented by Terra et al
29
are an impor-
tant preliminary step toward a better understanding of
interindividual variability in the response to
1
AR an-
tagonists in patients with heart failure during the dose-
escalation period. The number of patients studied was
small, and additional larger studies will be required to
better dene the clinical impact of ADRB1 and
CYP2D6 polymorphism in patients with heart failure
receiving -blockers.
The authors have identied no conict of interest.
References
1. Consensus recommendations for the management of
chronic heart failure. On behalf of the membership of the
advisory council to improve outcomes nationwide in
heart failure. Am J Cardiol 1999;83:1A-38A.
2. Effect of metoprolol CR/XL in chronic heart failure: Meto-
prolol CR/XL Randomised Intervention Trial in Congestive
Heart Failure (MERIT-HF). Lancet 1999;353:2001-7.
3. Krum H, Roecker EB, Mohacsi P, Rouleau JL, Tendera
M, Coats AJ, et al, Carvedilol Prospective Randomized
Cumulative Survival (COPERNICUS) Study Group. Ef-
fects of initiating carvedilol in patients with severe
chronic heart failure: results from the COPERNICUS
Study. JAMA 2003;289:712-8.
4. Leizorovicz A, Lechat P, Cucherat M, Bugnard F. Biso-
prolol for the treatment of chronic heart failure: a meta-
analysis on individual data of two placebo-controlled
studiesCIBIS and CIBIS II. Cardiac Insufciency
Bisoprolol Study. Am Heart J 2002;143:301-7.
5. Zugck C, Haunstetter A, Kruger C, Kell R, Schellberg D,
Kubler W, et al. Impact of beta-blocker treatment on the
prognostic value of currently used risk predictors in conges-
tive heart failure. J Am Coll Cardiol 2002;39:1615-22.
6. Sin DD, McAlister FA. The effects of beta-blockers on
morbidity and mortality in a population-based cohort of
11,942 elderly patients with heart failure. Am J Med 2002;
113:650-6.
7. Gottlieb SS, Fisher ML, Kjekshus J, Deedwania P,
Gullestad L, Vitovec J, et al. Tolerability of beta-blocker
initiation and titration in the Metoprolol CR/XL Ran-
domized Intervention Trial in Congestive Heart Failure
(MERIT-HF). Circulation 2002;105:1182-8.
8. Macdonald PS, Keogh AM, Aboyoun CL, Lund M,
Amor R, McCaffrey DJ. Tolerability and efcacy of
carvedilol in patients with New York Heart Association
class IV heart failure. J Am Coll Cardiol 1999;33:924-31.
9. Waagstein F, Bristow MR, Swedberg K, Camerini F,
Fowler MB, Silver MA, et al. Benecial effects of meto-
prolol in idiopathic dilated cardiomyopathy. Metoprolol
in Dilated Cardiomyopathy (MDC) Trial Study Group.
Lancet 1993;342:1441-6.
10. Baxter AJ, Spensley A, Hildreth A, Karimova G,
OConnell JE, Gray CS. Beta blockers in older persons
with heart failure: tolerability and impact on quality of
life. Heart 2002;88:611-4.
11. Wuttke H, Rau T, Heide R, Bergmann K, Bohm M, Weil
J, et al. Increased frequency of cytochrome P450 2D6
poor metabolizers among patients with metoprolol-
associated adverse effects. Clin Pharmacol Ther 2002;72:
429-37.
12. Sofowora GG, Dishy V, Muszkat M, Xie HG, Kim RB,
Harris PA, et al. A common beta1-adrenergic receptor
polymorphism (Arg389Gly) affects blood pressure re-
sponse to beta-blockade. Clin Pharmacol Ther 2003;73:
366-71.
13. Johnson JA, Zineh I, Puckett BJ, McGorray SP, Yarandi
HN, Pauly DF.
1
Adrenergic receptor polymorphisms
and antihypertensive response to metoprolol. Clin Phar-
macol Ther 2003;74:44-52.
14. Liu J, Liu ZQ, Tan ZR, Chen XP, Wang, LS, Zhou G,
et al. Gly389Arg polymorphism of beta1-adrenergic re-
ceptor is associated with the cardiovascular response to
metoprolol. Clin Pharmacol Ther 2003;74:372-9.
15. Zineh I, Beitelshees AL, Gaedigk A, Walker JR, Pauly
DF, Eberst K, et al. Pharmacokinetics and CYP2D6 ge-
notypes do not predict metoprolol adverse events or
efcacy in hypertension. Clin Pharmacol Ther 2004;76:
536-44.
16. Lennard MS, Silas JH, Freestone S, Ramsay LE, Tucker
GT, Woods HF. Oxidation phenotypea major determi-
nant of metoprolol metabolism and response. N Engl J Med
1982;307:1558-60.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):123-6 Pharmacogenetics and -blockers in heart failure 125
17. McGourty JC, Silas JH, Lennard MS, Tucker GT, Woods
HF. Metoprolol metabolism and debrisoquine oxidation
polymorphismpopulation and family studies. Br J Clin
Pharmacol 1985;20:555-66.
18. Hamelin BA, Bouayad A, Methot J, Jobin J, Desgagnes
P, Poirier P, et al. Signicant interaction between the
nonprescription antihistamine diphenhydramine and the
CYP2D6 substrate metoprolol in healthy men with high
or low CYP2D6 activity. Clin Pharmacol Ther 2000;67:
466-77.
19. Giessmann T, Modess C, Hecker U, Zschiesche M, Daz-
ert P, Kunert-Keil C, et al. CYP2D6 genotype and induc-
tion of intestinal drug transporters by rifampin predict
presystemic clearance of carvedilol in healthy subjects.
Clin Pharmacol Ther 2004;75:213-22.
20. Clack DWJ, Morgan AKW, Waal-Manning H. Adverse
effects from metoprolol are not generally associated with
oxidation status [letter]. Br J Clin Pharmacol 1984;18:
965-7.
21. Maqbool A, Hall AS, Ball SG, Balmforth AJ. Common
polymorphisms of beta1-adrenoceptor: identication and
rapid screening assay [letter]. Lancet 1999;353:897.
22. Moore JD, Mason DA, Green SA, Hsu J, Liggett SB. Racial
differences in the frequencies of cardiac beta(1)-adrenergic
receptor polymorphisms: analysis of c145AG and
c1165GC. Hum Mutat 1999;14:271.
23. Rathz DA, Brown KM, Kramer LA, Liggett SB. Amino
acid 49 polymorphisms of the human beta1-adrenergic
receptor affect agonist-promoted trafcking. J Cardio-
vasc Pharmacol 2002;39:155-60.
24. Mason DA, Moore JD, Green SA, Liggett SB. A gain-of-
function polymorphism in a G-protein coupling domain of
the human beta1-adrenergic receptor. J Biol Chem 1999;
274:12670-4.
25. Ginsburg R, Bristow MR, Billingham ME, Stinson EB,
Schroeder JS, Harrison DC. Study of the normal and
failing isolated human heart: decreased response of fail-
ing heart to isoproterenol. Am Heart J 1983;106:535-40.
26. Bristow MR, Ginsburg R, Minobe W, Cubiccioti RS,
Segman WS, Lurie K, et al. Decreased catecholamine
sensitivity and beta-adrenergic-receptor density in failing
human hearts. N Engl J Med 1982;307:205-11.
27. Heilbrunn SM, Shah P, Bristow MR, Valantine HA, Gins-
burg R, Fowler MB. Increased beta-receptor density and
improved hemodynamic response to catecholamine stimu-
lation during long-term metoprolol therapy in heart failure
from dilated cardiomyopathy. Circulation 1989;79:483-90.
28. Perez JM, Rathz DA, Petrashevskaya NN, Hahn HS,
Wagoner LE, Schwartz A, et al. Beta1-Adrenergic recep-
tor polymorphisms confer differential function and pre-
disposition to heart failure. Nat Med 2003;9:1300-5.
29. Terra SG, Pauly DF, Lee CR, Patterson JH, Adams KF
Jr, Schoeld RS, et al. -Adrenergic receptor polymor-
phisms and responses during titration of metoprolol con-
trolled release/extended release in heart failure. Clin
Pharmacol Ther 2005;77:127-37.
CLINICAL PHARMACOLOGY & THERAPEUTICS
126 Muszkat and Stein MARCH 2005
PHARMACOGENETICS AND
GENOMICS
-Adrenergic receptor polymorphisms and
responses during titration of metoprolol
controlled release/extended release in
heart failure
Objective: -Blockers require careful initiation and titration when used in patients with heart failure. Some
patients tolerate -blocker therapy initiation without difficulty, whereas in other patients this period presents
clinical challenges. We tested the hypothesis that polymorphisms at codons 389 (Arg389Gly) and 49
(Ser49Gly) of the
1
-adrenergic receptor would be associated with differences in initial tolerability of
-blocker therapy in patients with heart failure. We also tested whether polymorphisms in the
2
-adrenergic
receptor, G-protein
s
subunit (G
s
), and cytochrome P450 (CYP) 2D6 genes or S-metoprolol plasma
concentrations were associated with -blocker tolerability.
Methods: Sixty-one -blockernaive patients with systolic heart failure were prospectively enrolled. Patients
began taking 12.5 to 25 mg metoprolol controlled release/extended release with titration every 2 weeks (as
tolerated) to 200 mg/d or the maximum tolerated dose over a period of 8 to 10 weeks. Decompensation was
the composite of death, heart failure hospitalization, increase in other heart failure medications, or need to
discontinue metoprolol. End points were assessed during the titration period.
continued on next page
Steven G. Terra, PharmD, Daniel F. Pauly, MD, PhD, Craig R. Lee, PharmD,
J. Herbert Patterson, PharmD, Kirkwood F. Adams, Jr, MD,
Richard S. Schofield, MD, Bernadette S. Belgado, PharmD, Karen K. Hamilton, MD,
Juan M. Aranda, MD, James A. Hill, MD, MS, Hossein N. Yarandi, PhD,
Joseph R. Walker, PharmD, Michael S. Phillips, PhD, Craig A. Gelfand, PhD, and
Julie A. Johnson, PharmD Gainesville, Fla, Chapel Hill, NC, and Princeton, NJ
From the Department of Pharmacy Practice, College of Pharmacy,
Division of Cardiology, Department of Medicine, Department of
Veterans Affairs Medical Center, and College of Nursing and
Biostatistics Unit, University of Florida, Gainesville; Division of
Pharmacotherapy, University of North Carolina at Chapel Hill
School of Pharmacy, and Division of Cardiology, Department of
Medicine, University of North Carolina, Chapel Hill; and Orchid
Biosciences Inc, Princeton.
This work was supported by grants from Orchid Biosciences Inc, the
National Heart, Lung, and Blood Institute (HL68834), and by
General Clinical Research Centers program grants RR00082 (Uni-
versity of Florida) and RR00046 (University of North Carolina,
Chapel Hill), Division of Research Resources, National Institutes
of Health. Dr Terra was an American College of Clinical Pharmacy
Merck Cardiovascular Research Fellow and subsequently an
American Foundation for Pharmaceutical Education Clinical Phar-
macy Post-PharmD Fellow in Biomedical Research Sciences at the
time of this work.
Received for publication July 1, 2004; accepted Oct 1, 2004.
Reprint requests: Julie A. Johnson, PharmD, University of Florida
College of Pharmacy, Department of Pharmacy Practice, 1600 SW
Archer Rd, PO Box 100486, Gainesville, FL 32610.
E-mail: johnson@cop.u.edu
Clin Pharmacol Ther 2005;77:127-37.
0009-9236/$30.00
Copyright 2005 by the American Society for Clinical Pharmacology
and Therapeutics.
doi:10.1016/j.clpt.2004.10.006
127
Results: The overall rate of decompensation was not different between the codon 49 or 389 genotypes.
However, a significantly greater percentage of patients with the Gly389 variant required increases in heart
failure medications as compared with Arg389 homozygotes (48% versus 14%, respectively; P .006).
Similarly, patients with the Ser49 homozygous genotype were significantly more likely to require increases in
concomitant heart failure therapy as compared with Gly49 carriers (41% versus 11%, respectively; P .03).
Neither CYP2D6 genotypes nor metoprolol pharmacokinetics differed between patients with and those
without a decompensation event. There was no association between the
2
-adrenergic receptor or G
s
genes.
Statistical analysis
Power calculations were based on the assumption
that 15% of patients permanently discontinue -blocker
therapy and another 35% are unable to reach the target
dose (for a total of 50% with poor tolerability).
31
On the
basis of this and a 56% frequency of the Arg389 ho-
mozygous genotype, with 60 patients, we had 80%
power to detect a 2.4-fold relative risk of poor tolera-
bility among homozygous Arg389 patients as com-
pared with Gly389 carriers with a 2-tailed of .05. We
would similarly have 80% power to detect a 2.4-fold
relative risk of decompensation in Arg389 homozy-
gotes and Gly389 carriers. Patients were compared by
genotype relative to the decompensation endpoints by
use of a chi square or Fisher exact test, as appropriate.
Patients were also divided by genotype and compared
with regard to continuous parameters (eg, MLWHF
scores, 6-minute walk distance, dose of metoprolol
CR/XL at end of titration, systolic blood pressure, and
heart rate), as well as changes in these parameters.
Comparisons were made by unpaired t test, ANOVA,
or Kruskal-Wallis test, as appropriate. Discrete vari-
Table I. Baseline demographic characteristics stratied by
1
AR genotype
Variable
Arg389Arg
(n 28)
Gly389 carriers
(n 33)
Ser49Ser
(n 43)
Gly49 carriers
(n 17)
Age (y) 59 12 55 14 58 13 55 12
Men 18 (64%) 19 (57%) 25 (58%) 11 (64%)
White/black/Hispanic 21 (75%)/7 21 (63%)/11 32 (74%)/11 10 (59%)/6
(25%)/0 (0%) (34%)/1 (3%) (26%)/0 (0%) (35%)/1 (6%)
NYHA functional class II/III 17 (60%)/11 (40%) 22 (66%)/11 (34%) 25 (58%)/18 (42%) 13 (76%)/4 (23%)
Ischemic heart failure 7 (25%) 12 (36%) 15 (34%) 4 (23%)
Body mass index (kg/m
2
) 27 6 29 6 27 6 30 6
Left ventricular ejection fraction (%) 23 5 22 9 21 7 25 9
Serum sodium (mEq/L) 138 3 138 3 138 4 139 3
Concomitant medical history
Diabetes 7 (25%) 12 (36%) 14 (32%) 5 (29%)
Hypertension 14 (50%) 17 (51%) 21 (48%) 9 (52%)
Atrial brillation 6 (21%) 6 (18%) 11 (25%) 1 (5%)
Background therapy
ACE inhibitor/ARB 28 (100%) 33 (100%) 43 (100%) 17 (100%)
Furosemide 22 (78%) 31 (93%) 38 (88%) 14 (82%)
Digoxin 18 (64%) 28 (84%) 33 (76%) 13 (76%)
Spironolactone 9 (32%) 8 (24%) 12 (28%) 5 (29%)
Data are given as mean SD or number and percent. There were no signicant differences between genotype groups for any parameter.
1
AR,
1
-Adrenergic receptor; NYHA, New York Heart Association; ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker.
CLINICAL PHARMACOLOGY & THERAPEUTICS
130 Terra et al MARCH 2005
ables are listed as frequency counts and percentages,
whereas continuous variables are described as mean
SD or median and interquartile range. Polymorphisms
that resulted in a homozygous variant genotype fre-
quency of less than 2% were combined with the het-
erozygotes for analysis.
All statistical analyses were performed with the use
of SAS (version 8; SAS Institute, Cary, NC). A P value
of less than .05 was considered statistically signicant.
RESULTS
Table I shows the clinical characteristics of the 61
patients enrolled in this study, with data stratied by the
1
AR genotypes at codons 49 and 389 because these
polymorphisms were the primary focus of the study.
There was 1 subject in whom determination of the
codon 49 genotype was unsuccessful. The Arg389 al-
lele frequency was 0.72; there were 32 heterozygotes
and 1 Gly389 homozygote. The Ser49 allele frequency
was 0.85; there were 17 heterozygotes and no Gly49
homozygotes. None of the baseline characteristics
listed in Table I was statistically signicant between
genotype groups. More Gly389 carriers were receiving
digoxin and furosemide at baseline compared with
Arg389 homozygotes, but the difference was not sta-
tistically signicant (P .07 and P .12, respec-
tively). Patients were predominantly middle-aged white
men; approximately 35% of patients had NYHA func-
tional class III heart failure at enrollment, and mean
ejection fraction was 22% 7%.
Impact of
1
AR polymorphisms on poor
tolerability to metoprolol CR/XL
A similar percentage of patients in each
1
AR ge-
notype was unable to reach the target dose of metopro-
lol CR/XL or required discontinuation of therapy dur-
ing the titration of therapy. Overall, 19 of 28 Arg389
homozygous patients (67%) had poor tolerability to
metoprolol CR/XL compared with 25 of 33 Gly389
carriers (75%) (P .62). For the codon 49 polymor-
phism, the frequency of poor tolerability was not dif-
ferent between genotype groups (72% and 76% for
Ser49Ser and Gly49 carriers, respectively).
Impact of
1
AR polymorphisms on risk of
decompensated heart failure
Table II lists the end points for decompensated heart
failure for the codon 49 and 389 genotype groups.
There was a trend for Gly389 carriers to have a higher
incidence of the composite end point of decompensated
heart failure compared with Arg389 homozygotes
within 8 to 10 weeks of initiation of metoprolol CR/XL.
This was largely the result of a signicantly greater
percentage of Gly389 carriers requiring increases in
other heart failure medications (mostly diuretics) for
symptoms of worsening heart failure during -blocker
titration (48% versus 14%, respectively; P .006).
Although not statistically signicant, patients with the
Ser49 homozygous genotype had a nominally higher
rate of decompensation compared with Gly49 carriers.
This nding was driven largely by a signicantly higher
proportion requiring increases in heart failure medica-
tions to treat worsening symptoms. The nal dose of
metoprolol CR/XL was not signicantly different be-
tween patients who required increases in heart failure
medications compared with those who did not reach
this end point (90 mg/d versus 118 mg/d, respectively;
P .16). There were also no signicant differences in
the baseline characteristics of age, serum sodium level,
furosemide dose, 6-minute walk distance, scores on the
Table II. Risk of decompensated heart failure stratied by
1
AR genotypes
Arg389Arg
(n 28)
Gly389 carriers
(n 33) P value
Ser49Ser
(n 43)
Gly49 carriers
(n 17) P value
Decompensated heart failure
(composite)
8 (28%) 16 (48%) .11 19 (44%) 5 (29%) .29
Death 0 0 0 0
Hospitalization for worsening
heart failure
1 (3.5%) 2 (6%) .99 3 (6%) 0
Increase of heart failure
medications
4 (14%) 16 (48%) .006 18 (41%) 2 (11%) .03
Discontinuation of metoprolol
CR/XL
3 (10%) 3 (9%) .99 3 (6%) 3 (17%) .33
Data are given as number and percent.
CR/XL, Controlled release/extended release.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):127-37 -Receptor polymorphisms and response to metoprolol 131
MLWHF questionnaire, heart rate, or systolic blood
pressure between patients who required increased heart
failure medications and those who did not reach this
end point (data not shown).
1
AR diplotypes (based on codon 49 and 389
polymorphisms) were determined in 60 patients. The
observed diplotypes were SR/SG (Ser49Ser plus
Arg389Gly) (n 25), SR/SR (Ser49Ser plus Arg389Arg)
(n 17), SR/GR (Ser49Gly plus Arg389Arg) (n
10), SG/GR (Ser49Gly plus Arg389Gly) (n 7), and
SG/SG (Ser49Ser plus Gly389Gly) (n 1). Fig 1
shows that there was a signicant difference in the
percentage of patients who required increases in
heart failure medications when the response was
stratied by the 4 most commonly observed
1
AR
diplotypes (P .01). The presence of a Gly389
variant was associated with an increased likelihood
of requiring increases in concomitant heart failure
Fig 1. Percentage of patients requiring increases in concomitant heart failure medications during
metoprolol CR/XL titration according to
1
-adrenergic receptor (
1
AR) diplotype.
Table III. Impact of
1
AR genotypes on 6-minute walk distance, quality-of-life scores, hemodynamics, and dose
of metoprolol CR/XL at end of titration
Arg389Arg
(n 28)
Gly389 carriers
(n 33)
Ser49Ser
(n 43)
Gly49 carriers
(n 17)
Baseline
End of
titration Baseline
End of
titration Baseline
End of
titration Baseline
End of
titration
Metoprolol CR/XL
dose (mg/d)
119 72 100 74 107 71 108 79
6-min walk
distance (m)
330 108 357 127 301 75 293 102* 298 89 306 111 346 88 359 131
MLWHF 28 21 21 22 36 27 30 20 31 25 25 20 37 25 31 25
Heart rate (beats/min) 80 12 64 12 79 10 70 12 80 11 68 12 79 12 65 12
Systolic blood
pressure (mm Hg)
120 18 110 20 119 17 111 18 118 18 110 18 121 17 111 21
Data are given as mean SD.
MLWHF, Minnesota Living with Heart Failure questionnaire.
*P .06, versus Arg389Arg at end of titration.
P .07, versus Ser49Ser at baseline.
P .09, versus Arg389Arg at end of titration.
CLINICAL PHARMACOLOGY & THERAPEUTICS
132 Terra et al MARCH 2005
therapy. In all, 52% of patients with the SR/SG
diplotype (Ser49Ser plus Arg389Gly) and 42% of
patients with the SG/GR diplotype (Ser49Gly plus
Arg389Gly) required increases in heart failure med-
ications during metoprolol CR/XL titration. The
composite decompensation end point was not signif-
icantly different between groups (52%, 29%, 30%,
and 42% for the SR/SG, SR/SR, SR/GR, and SG/GR
diplotypes, respectively; P .46).
Impact of
1
AR polymorphisms on 6-minute walk
distance, quality of life, and dose of metoprolol
CR/XL attained
We also assessed tolerability to metoprolol CR/XL
by measuring submaximal exercise tolerance (6-minute
walk distance) scores on the MLWHF questionnaire
and dose of metoprolol CR/XL achieved at the end of
the titration period. Table III lists these data for the
1
AR polymorphisms. At baseline, there were no sig-
nicant differences in any of these parameters between
the codon 49 and codon 389 genotypes. However,
Gly49 carriers tended to cover a greater distance in the
baseline 6-minute walk test compared with Ser49 ho-
mozygous patients. At the end of the titration period,
there was a trend for Arg389 homozygous patients to
have a greater exercise tolerance compared with
Gly389 carriers. The change from baseline in the
6-minute walk distance was not signicantly different
between the codon 389 genotype groups. Scores on the
MLWHF questionnaire and reduction in heart rate and
systolic blood pressure were similar between
1
AR
genotype groups (Table III). There were no differences
in metoprolol CR/XL dose at the end of the titration
period between genotype groups (Table III). In all, 67%
of patients with the Arg389Arg genotype attained
metoprolol CR/XL doses of 100 mg/d or greater com-
pared with 51% of Gly 389 carriers (P .30). For
codon 49, a similar percentage of patients reached a
nal dose of 100 mg/d or greater (58% in each codon
49 group).
Impact of
2
AR and G
s
polymorphisms on risk
of decompensation and tolerability to metoprolol
CR/XL
The
2
AR and G
s
polymorphisms were not associ-
ated with the dose of metoprolol CR/XL attained at the
end of the titration period. For example, among the 3
codon 16
2
AR genotypes, the dose of metoprolol was
essentially identical (107, 109, and 108 mg/d for
Arg16Arg, Arg16Gly, and Gly16Gly, respectively).
Other measures of tolerability such as 6-minute walk
distances or quality-of-life assessment were also not
different between genotype groups (data not shown).
The risk of development of decompensated heart failure
was also not different among the various genotypes
within these genes. For example, decompensated heart
failure occurred among 40% of Glu27 carriers com-
pared with 37% of patients who were Gln27 homozy-
gotes, and approximately 30% of patients with each
genotype required increased doses of heart failure med-
ication.
Metoprolol pharmacokinetics and CYP2D6
genotypedetermined phenotype
There were no pharmacokinetic differences between
patients with and those without a decompensation event
at any dose of metoprolol CR/XL. The medians and
interquartile ranges of S-metoprolol concentrations for
patients with decompensation were 4.8 ng/mL (4.0-7.7
ng/mL), 11.9 ng/mL (5.7-19.2 ng/mL), 26.8 ng/mL
(15.4-35.4 ng/mL), and 46.0 ng/mL (31.7-85 ng/mL)
for the 25-, 50-, 100-, and 200-mg doses, respectively.
Table IV. S-metoprolol plasma concentrations and decompensation rate stratied by CYP2D6 phenotype
EM IM PM P value*
Dose
25 mg 4.0 (2.8-7.3) (n27) 7.6 (4.0-10.4) (n8) 15.6 (12.9-20.2) (n4) .005
50 mg 11.9 (4.5-15.1) (n27) 6.4 (3.6-19.7) (n7) 27.0 (25.6-35.9) (n4) .01
100 mg 17.7 (11.9-28.5) (n21) 42.5 (13.3-42.9) (n5) 60.2 (49.2-83.2) (n3) .03
200 mg 50.3 (31.2-56.9) (n9) 69.5 (65.8-119.6) (n3) 169.6 (146.6-192.5) (n2) .02
Decompensation rate 32% 50% 25% .61
Data are given as median and interquartile range or percentage.
EM, Extensive metabolizer; IM, intermediate metabolizer; PM, poor metabolizer; n, sample size.
*P value between groups.
P .02, versus IM.
P .001, versus EM.
P .04, versus IM.
P .004, versus EM.
P .03, versus EM.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):127-37 -Receptor polymorphisms and response to metoprolol 133
For patients with no decompensation, the medians and
interquartile ranges of concentrations were 5.7 ng/mL
(3.1-9.4 ng/mL), 12.1 ng/mL (4.8-20.1 ng/mL), 22.9
ng/mL (11.6-39.1 ng/mL), and 56.5 ng/mL (36.0-69.3
ng/mL) at these doses, respectively. Likewise, there
were no differences in S-metoprolol concentrations be-
tween the codon 389 genotype groups for all meto-
prolol CR/XL doses. For example, the median
S-metoprolol concentrations in Arg389 homozygotes
and Gly389 carriers at 100 mg/d were 22.8 ng/mL
(interquartile range, 13.3-35.3 ng/mL) and 22.6 ng/mL
(interquartile range, 12-41.6 ng/mL), respectively.
CYP2D6 genotype data were available in 51 pa-
tients, with genotype-determined phenotype distribu-
tions of 37 EMs, 10 IMs, and 4 PMs. As expected,
S-metoprolol plasma concentrations were signicantly
different between CYP2D6 phenotypes (Table IV). It
should be noted that plasma concentrations were mea-
sured in each patient at each dose that they received.
Therefore patients reaching the 200-mg dose would
have data included in each of the 4 dose steps. Consis-
tent with the plasma concentration data, the risk of
cardiac decompensation was not inuenced by
CYP2D6 genotypedetermined phenotype (32%, 50%,
and 25% for EMs, IMs, and PMs, respectively; P
.61). Similarly, the need for increases in background
heart failure medications was not different between the
3 CYP2D6 groups (data not shown). The nal daily
metoprolol CR/XL doses were 106 69, 100 75,
and 150 57 mg/d for EMs, IMs, and PMs (P .45).
There were no signicant differences in baseline or
end-of-study heart rate or systolic blood pressure values
when stratied by CYP2D6 genotypedetermined phe-
notype.
DISCUSSION
To our knowledge, this is the rst study that has
investigated the relationship between early tolerability
to a -blocker and genotype. The major nding of this
study was that the heart failure patients with the Gly389
variant and those with the Ser49 homozygous genotype
were signicantly more likely to require increases in
heart failure medications during -blocker titration than
patients with other
1
AR genotypes. Furthermore, the
combination of the codon 49 and 389 polymorphisms
(ie,
1
AR diplotypes) was also associated with a sig-
nicantly higher risk for increases in concomitant heart
failure medications, suggesting that these patients may
require more frequent follow-up during titration. In
support of the clinical end points, patients who were
Gly389 carriers tended to have depressed submaximal
exercise performance compared with Arg389 homozy-
gotes at the end of the titration period. Similarly, at
baseline, there was a trend for Gly49 carriers to have a
greater exercise performance compared with Ser49 ho-
mozygous patients. The composite end point of decom-
pensated heart failure tended to be higher among
Gly389 carriers as compared with Arg389 homozy-
gotes, but this study was slightly underpowered to
detect a difference in the magnitude observed. Specif-
ically, we had 80% power to detect a 2.4-fold differ-
ence in decompensation rates but observed a 1.7-fold
difference between Arg389 homozygotes and Gly389
carriers. Thus it is possible that there would be signif-
icant differences in the decompensation end point with
a larger sample.
The 6-minute walk distance is frequently used in
clinical trials as a measurement of submaximal exercise
performance, and exercise capacity is directly related to
survival rate among heart failure patients.
32
Our nd-
ings on submaximal exercise tolerance are consistent
with a previous study in which peak myocardial oxygen
consumption was signicantly lower among patients
with the Gly389 variant and Ser49 homozygotes.
33
The
nding of reduced submaximal and maximal exercise
tolerance among Gly389 carriers is also entirely con-
sistent with results of in vitro studies demonstrating that
the Gly389 variant is hypofunctional.
4
Moreover, in
one study heart failure patients with the Gly49 poly-
morphism had an improved survival rate as compared
with those who were Ser49 homozygotes.
15
It is unclear
whether the differences in exercise performance ac-
count for any of the apparent survival benet. The
1
AR polymorphisms at codons 49 and 389 are no
more common in heart failure patients than in a control
population.
15,34
We did not observe differences by genotype in the
heart rate or blood pressure response to metoprolol.
Among a number of studies looking at the association
of heart rate with codon 389 genotype, the vast majority
have shown no association. This is independent of
whether the study investigated the association with
resting heart rate,
7,8
the heart rate response to exer-
cise,
7,8
or the heart rate response to -blockade.
9,10
It is
unclear why heart rate or the heart rate response to
drugs is not associated with the codon 389 genotype
when other drug response phenotypes (eg, ejection
fraction change, blood pressure response to
-blockade) have been fairly consistently associated
with codon 389 genotype.
9-12
However, there is little
doubt that the processes controlling heart rate and blood
pressure, for example, are different, and thus it is not
unreasonable that genetic associations might be differ-
ent for different phenotype responses to an agonist
CLINICAL PHARMACOLOGY & THERAPEUTICS
134 Terra et al MARCH 2005
versus an antagonist. The blood pressure ndings are in
contrast to previous ndings from our laboratory
9
and
others regarding the genetic associations with
-blocker effect on blood pressure. However, this is not
surprising for a number of reasons, and these are ex-
plained relative to our previous ndings in hypertensive
patients.
The rst factor is the method of measuring blood
pressure. In this study blood pressure assessments were
based on duplicate values obtained in the clinic,
whereas in our previous study we used 24-hour ambu-
latory blood pressure. The latter approach signicantly
reduces variability and consequently increases study
power. The second factor is the study population. Pa-
tients with heart failure are known to have diminished
blood pressurelowering effects from -blockers and
other antihypertensive drugs. This is presumably a re-
sult of the heightened activity of the sympathetic and
renin-angiotensin system that occurs in patients with
heart failure and the interplay between vascular resis-
tance and stroke volume, which differs dramatically
between those with normal (eg, hypertensive) versus
abnormal left ventricular function. Third, in the hyper-
tension study metoprolol was titrated to a goal diastolic
blood pressure response and the mean doses achieved
were substantially higher than those in the population
of heart failure patients. Fourth, patients in this study
received a variety of medications known to affect blood
pressure. This is in contrast to the earlier study in
hypertensive patients, in which metoprolol was the only
prescription medication allowed. We believe these dif-
ferences limit any comparison to our previous study of
hypertensive patients
The 2 most commonly used -blockers in heart fail-
ure, metoprolol and carvedilol, are metabolized by the
polymorphic CYP2D6 enzyme. The CYP2D6 enzyme
accounts for 70% to 80% of the metabolism of meto-
prolol.
35
There have been suggestions that one potential
cause of initial poor tolerability might be excessively
high concentrations in those who are CYP2D6 poor
metabolizers,
36
although to our knowledge this has
never been documented. In this study there were no
pharmacokinetic differences between patients with de-
compensation versus those without decompensation
during the titration period. Thus, consistent with data
regarding hypertension from our laboratory,
37
we do
not have any evidence that S-metoprolol concentration
contributes importantly to the tolerability of metopro-
lol. This nding is perhaps surprising but suggests that
there are substantial differences in the sensitivity to
-blockade (ie, pharmacodynamics) that may be more
important in determining the level of -blockade at a
given concentration.
38
Polymorphisms within the coding region of
2
AR
have been shown to be associated with altered exercise
tolerance among heart failure patients.
39
The Glu27
allele in the
2
AR gene was also shown to be associ-
ated with improved left ventricular ejection fraction
after treatment with carvedilol.
18
However, in this
study we did not observe any differences in either
baseline or end-of-study 6-minute walk distance ac-
cording to the
2
AR genotypes. Similarly, the risk of
decompensated heart failure was similar for polymor-
phisms in
2
AR. There are several explanations for the
differences in ndings between our study and that of
carvedilol pharmacogenetics in heart failure. First,
carvedilol is a nonselective -blocker (and thus blocks
the
2
-receptor) and so would be expected to be more
likely to exhibit a signicant association with the
2
-
receptor polymorphisms than with the
1
-selective
metoprolol. In addition, the end points in the 2 studies
were quite different, with a focus on short-term toler-
ability in this study and a focus on long-term ventric-
ular changes in the carvedilol study. Thus it would
probably be inappropriate to view the ndings from
these 2 studies as inconsistent, because there were
important differences in the study designs.
Although initiation of -blocker therapy can cause
worsening heart failure, analyses from clinical trials
failed to demonstrate any increased risk of cardiac
deterioration as compared with placebo during the ti-
tration period.
3,40
These clinical trial data are likely
explained by those patients who had initial worsening
of heart failure symptoms with -blocker initiation
being counterbalanced by those deriving early benet
from the -blocker. Consensus guidelines recommend
that all patients with systolic heart failure and without
contraindications be treated with a -blocker, and most
can be maintained on some -blocker dose. It is rec-
ognized that there are certain challenges associated
with the initiation of -blocker therapy in heart failure,
and known clinical and demographic risk factors that
are associated with worsening heart failure
31,41
should
be taken into account and titration individualized on the
basis of the clinical condition of the patient. In the
future, genetic information may also be helpful in in-
dividualizing the titration plan.
The authors have no conicts of interest to disclose.
References
1. Hunt SA, Baker DW, Chin MH, Cinquegrani MP, Feld-
man AM, Grancis GS, et al. ACC/AHA guidelines for the
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):127-37 -Receptor polymorphisms and response to metoprolol 135
evaluation and management of chronic heart failure in the
adult: executive summary. A report of the American
College of Cardiology/American Heart Association Task
Force on Practice Guidelines (Committee to revise the
1995 Guidelines for the Evaluation and Management of
Heart Failure). J Am Coll Cardiol 2001;38:2101-13.
2. Cleland JGF, Cohen-Solal A, Aguilar JC, Dietz R, East-
augh J, Follath F, et al, for the IMPROVEMENT of Heart
Failure Programme Committees and Investigators. Man-
agement of heart failure in primary care (the IMPROVE-
MENT of Heart Failure Programme): an international
survey. Lancet 2002;360:1631-9.
3. Gottlieb SS, Fischer ML, Kjekshus J, Deedwania P,
Gullestad L, Vitovec J, et al. MERIT-HF Investigators.
Tolerability of beta-blocker initiation and titration in the
Metoprolol CR/XL Randomised Intervention Trial in
Congestive Heart Failure (MERIT-HF). Circulation
2002;105:1182-8.
4. Mason DA, Moore JD, Green SA, Liggett SB. A gain-
of-function polymorphism in a G-protein coupling do-
main of the human
1
-adrenergic receptor. J Biol Chem
1999;274:12670-4.
5. Sandilands AJ, OShaughnessy KM, Brown MJ. Greater
inotropic and cyclic AMP responses evoked by noradren-
aline through Arg389
1
-adrenoceptors versus Gly389
1
-adrenoceptors in isolated human atrial myocardium.
Br J Pharmacol 2003;138:386-92.
6. Molenaar P, Rabnott G, Yang I, Fong KM, Savarimuthu
SM, Li L, et al. Conservation of the cardiostimulant
effects of (-)-norepinephrine across Ser49Gly and
Gly389Arg beta
1
-adrenergic receptor polymorphisms in
human right atrium in vitro. J Am Coll Cardiol 2002;40:
1275-82.
7. Xie HG, Dishy V, Sofowora G, Kim, RB, Landau R,
Smiley RM, et al. Arg
389
Gly
1
-adrenoceptor polymor-
phism varies in frequency among different ethnic groups
but does not alter response in vivo. Pharmacogenetics
2001;11:191-7.
8. Buscher R, Belger H, Eilmes KJ, Tellkamp R, Radke J,
Dhein S, et al. In-vivo studies do not support a major
functional role for the Gly
389
Arg
1
-adrenoceptor poly-
morphism in humans. Pharmacogenetics 2001;11:199-
205.
9. Johnson JA, Zineh I, Puckett BJ, McGorray SP, Yarandi
HN, Pauly DF.
1
-Adrenergic receptor polymorphisms
and antihypertensive response to metoprolol. Clin Phar-
macol Ther 2003;74:44-52.
10. Sofowora GG, Dishy V, Muszkat M, Xie HG, Kim RB,
Harris PA, et al. A common
1
-adrenergic receptor poly-
morphism (Arg389Gly) affects blood pressure response
to -blockade. Clin Pharmacol Ther 2003;73:366-71.
11. Liu J, Liu ZQ, Tan ZR, Chen XP, Wang LS, Zhou G, et
al. Gly389Arg polymorphism of
1
-adrenergic receptor
is associated with the cardiovascular response to meto-
prolol. Clin Pharmacol Ther 2003;74:372-9.
12. Mialet Perez J, Rathz DA, Petrashevskaya NN, Hahn HS,
Wagoner LE, Schwartz A, et al. Beta 1-adrenergic recep-
tor polymorphisms confer differential function and pre-
disposition to heart failure. Nat Med 2003;9:1300-5.
13. Rathz DA, Brown KM, Kramer LA, Liggett SB. Amino
acid polymorphism of the human beta 1-adrenergic re-
ceptor affect agonist-promoted trafcking. J Cardiovasc
Pharmacol 2002;39:155-60.
14. Levin MC, Marullo S, Muntaner O, Andersson B, Mag-
nusson Y. The myocardium-protective Gly-49 variant of
the
1
-adrenergic receptor exhibits constitutive activity
and increased desensitization and down-regulation. J Biol
Chem 2002;277:30429-35.
15. Borjesson M, Magnusson Y, Hjalmarson Y, Andersson
B. A novel polymorphism in the gene coding for the
beta(1)-adrenergic receptor associated with survival in
patients with heart failure. Eur Heart J 2000;21:1853-8.
16. Dishy V, Sofowora GG, Xie HG, Kim RB, Byrne DW,
Stein CM, et al. The effect of common polymorphisms of
the beta-2-adrenergic receptor on agonist-mediated vas-
cular desensitization. N Engl J Med 2001;345:1030-5.
17. Jia H, Hingorani AD, Sharma P, Hopper R, Dickerson C,
Trutwein D, et al. Association of the G(s) alpha gene with
essential hypertension and response to beta-blockade.
Hypertension 1999;34:8-14.
18. Kaye DM, Smirk B, Williams C, Jennings G, Esler M,
Holst D. Beta-adrenoceptor genotype inuences the re-
sponse to carvedilol in patients with congestive heart
failure. Pharmacogenetics 2003;13:379-82.
19. Guyatt GH, Sullivan MJ, Thompson PJ, Fallen EL, Pugs-
ley SO, Taylor DW, et al. The 6-minute walk: a new
measure of exercise capacity in patients with chronic
heart failure. Can Med Assoc J 1985;132:919-23.
20. Rector TS, Cohn JN. Assessment of patient outcome with
the Minnesota Living with Heart Failure questionnaire:
reliability and validity during a randomized, double-
blind, placebo-controlled trial of pimobendan. Pimoben-
dan Multicenter Research Group. Am Heart J 1992;124:
1017-25.
21. Johnson JA, Burlew BS. Metoprolol metabolism via cy-
tochrome P4502D6 in ethnic populations. Drug Metab
Dispos 1996;24:350-5.
22. Terra SG, McGorray SP, Wallace MR, Picoult-Newberg
L, Pepine CJ, Johnson JA. Linkage disequilibrium of
common beta-1 adrenergic receptor polymorphisms [ab-
stract]. Clin Pharmacol Ther 2002;71:P70.
23. Bell PA, Chaturvedi S, Gelfand CA, Huang CY, Koch-
ersperger M, Kopla R, et al. SNPstream UHT: ultra-high
throughput SNP genotyping for pharmacogenomics and
drug discovery. Biotechniques 2002;Suppl:70-2, 74,
76-7.
24. Kimura S, Umeno M, Skoda RC, Meyer UA, Gonzalez
FJ. The human debrisoquine 4-hydroxylase (CYP2D)
locus: sequence and identication of the polymorphic
CYP2D6 gene, a related gene, and a pseudogene. Am J
Hum Genet 1989;45:889-904.
CLINICAL PHARMACOLOGY & THERAPEUTICS
136 Terra et al MARCH 2005
25. Gaedigk A, Gotschall RR, Forbes NS, Simon SD, Leeder
JS. Optimization of cytochrome P4502D6 (CYP2D6)
phenotype assignment using a genotyping algorithm
based on allele frequency data. Pharmacogenetics 1999;
9:669-82.
26. Griese EU, Zanger UM, Brudermanns U, Gaedigk A,
Mikus G, Morike K, et al. Assessment of the predictive
power of genotypes for the in-vivo catalytic function of
CYP2D6 in a German population. Pharmacogenetics
1998;8:15-26.
27. Marcucci KA, Pearce RE, Crespi C, Leeder JS, Gaedigk A.
Characterization of cytochrome P450 2D6.1 (CYP2D6.1),
CYP2D6.2 and CYP2D6.17 activities toward model
CYP2D6 substrates dextromethorphan, bufuralol, and de-
brisoquine. Drug Metab Dispos 2002;30:1-7.
28. Raimundo S, Fischer J, Eichelbaum M, Griese EU,
Schwab M, Zanger UM. Elucidation of the genetic basis
of the common intermediate metabolizer phenotype for
drug oxidation by CYP2D6. Pharmacogenetics 2000;10:
577-81.
29. Oscarson M, Hidestrand M, Johansson I, Ingelman-
Sundberg M. A combination of mutations in the
CYP2D6*17 (CYP2D6Z) allele causes alterations in en-
zyme function. Mol Pharmacol 1997;52:1034-40.
30. Sachse C, Brockmoller J, Bauer S, Roots I. Cytochrome
P450 2D6 variants in a Caucasian population: allele
frequencies and phenotypic consequences. Am J Hum
Genet 1997;60:265-71.
31. Anthonio RL, van Veldhuisen DJ, Breekland A, Crijns
HJGM, van Gilst WH. Beta-blocker titration failure is
independent of severity of heart failure. Am J Cardiol
2000;85:509-12.
32. Mancini DM, Eisen H, Kussmaul W, Mull R, Edmunds
LH Jr, Wilson JR. Value of peak exercise oxygen con-
sumption for optimal timing of cardiac transplantation in
ambulatory patients with heart failure. Circ Res 1991;83:
778-86.
33. Wagoner LE, Craft LL, Zengel P, McGuire N, Rathz DA,
Dorn GW II, et al. Polymorphisms of the
1
-adrenergic
receptor predict exercise capacity in heart failure. Am
Heart J 2002;144:840-6.
34. Tesson F, Charron P, Peuchmaurd M, Nicaud V, Cam-
bien F, Tiret L, et al. Characterization of a unique genetic
variant in the
1
-adrenoceptor gene and evaluation of its
role in idiopathic dilated cardiomyopathy. J Mol Cell
Cardiol 1999;31:1025-32.
35. Regardh CG, Johnsson G. Clinical pharmacokinetics of
metoprolol. Clin Pharmacokinet 1980;5:557-69.
36. Meadowcroft AM, Williamson KM, Patterson JH, Pieper
JA. Pharmacogenetics and heart failure: a convergence
with carvedilol. Pharmacotherapy 1997;17:637-9.
37. Beitelshees AL, Zineh I, Gaedigk A, Leeder JS, Walker
JR, Eberst K, et al. Effect of CYP2D6 phenotype and
metoprolol pharmacokinetics on adverse effects [ab-
stract]. Clin Pharmacol Ther 2004;75:P94.
38. Sowinski KM, Burlew BS, Johnson JA. Racial differ-
ences in sensitivity to the negative chronotropic effects of
propranolol in healthy males. Clin Pharmacol Ther 1995;
57:678-83.
39. Wagoner LE, Craft LL, Singh B, Suresh DP, Zengel PW,
McGuire N, et al. Polymorphisms of the
2
-adrenergic
receptor determine exercise capacity in patients with
heart failure. Circ Res 2000;86:834-40.
40. Krum H, Roecker EB, Mohacsi P, Rouleau JL, Tendera
M, Coats AJ, et al. Carvedilol Prospective Randomized
Cumulative Survival (COPERNICUS) Study Group. Ef-
fects of initiating carvedilol in patients with severe
chronic heart failure: results from the COPERNICUS
Study. JAMA 2003;289:712-8.
41. Schleman KA, Lindenfeld JA, Lowes BD, Bristow MR,
Ferguson D, Wolfel EE, et al. Predicting response to
carvedilol for the treatment of heart failure: a multivariate
retrospective analysis. J Card Fail 2001;7:4-12.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):127-37 -Receptor polymorphisms and response to metoprolol 137
Polymorphism screening in the cardiac K
act
(mV)
act fast
(ms)
act slow
(ms)
deact fast
(ms)
deact slow
(ms)
Block by
10 -mol/L
quinidine (%)
Wild type 389 49 5.50 2.13 1.26 0.26 6.58 1.88 14.89 1.19 62.31 6.16 49.6 7.5
A115V 402 19 6.11 3.23 1.25 0.25 7.37 1.20 13.32 2.45 60.85 4.50 47.9 8.3
A251T 413 59 4.83 1.40 1.40 0.26 7.39 0.82 16.29 2.71 73.28 13.17 47.9 6.0
P307S 407 43 6.78 2.08 1.07 0.25 5.58 0.62 14.00 3.41 57.16 2.29 47.8 3.5
P310L 381 48 8.63 2.06 1.91 0.31 6.48 1.56 13.05 6.39 71.77 6.28 54.3 2.4
P532L 400 44 6.01 2.81 1.39 0.46 6.33 1.40 5.03 2.87* 42.95 11.93 7.3 2.1
R578K 406 56 3.59 1.20 1.22 0.28 7.15 1.10 12.60 1.53 47.48 3.88 16.1 2.0
act fast and slow were measured by use of a depolarizing pulse to 50 mV. deact fast and slow were measured from the tail current obtained at 30 mV, after
a depolarizing pulse to 50 mV. Act current, Steady-state current measured at end of depolarizing pulse to 50 mV; V
)/k]), where I
max
is the maximal
current and k represents the slope factor. Electrophysi-
ologic data are presented as means SEM. Differences
between wild-type and mutant currents were analyzed
with a 1-way ANOVA. A P value .05 was consid-
ered statistically signicant.
RESULTS
KCNA5. Five synonymous and 6 nonsynonymous
polymorphisms, all with minor allele frequencies lower
than 7%, were found in KCNA5 (Fig 1 and Table I). No
polymorphism was present in all 3 populations.
Electrophysiology. Table II and Fig 2 show that I
Kur
generated by the KCNA5 variants A115V, A251T,
P307S, and P310L were nearly indistinguishable from
wild type. In addition, 10 mol/L of the prototypical
blocker quinidine produced approximately 50% inhibi-
tion of wild-type current at 50 mV, as previously
reported,
17
and the extent of inhibition was similar with
these 4 variants.
The gating of the remaining 2 variants, P532L and
R578K (both in the C-terminus), displayed only minor
differences compared with wild type (Fig 3 and Table
II). P532L did display slightly faster deactivation ki-
netics, with a fast time constant () of 5.0 2.9 ms
versus 14.9 1.2 ms for wild type (P .002). R578K
displayed a 9-mV positive shift in the voltage of half-
activation (V
4
knockout mice demonstrate increased extracellular
levels of dopamine.
44
Other studies have suggested that
nicotinic acetylcholine receptors containing only
7
subunits are involved in nicotine-induced dopamine
release,
45
mediate effects of the nicotine withdrawal
syndrome,
46
and are associated with risk for smoking.
26
Genetic differences within subunits may contribute to
altered smoking risk, although these effects are still
unclear.
47,48
Smokers modulate their smoking to maintain brain
nicotine levels within a certain concentration range,
49
and factors that alter nicotine clearance affect smoking
behavior.
50
It has been established in a previous study
that craving for cigarettes increases as levels of plasma
nicotine and breath carbon monoxide, an indicator of
smoke exposure, decline.
51
In this study, participants
were asked to smoke as desired until they were satis-
ed. Individuals who eliminated nicotine rapidly were
less likely to achieve clearly low craving scores even
after smoking freely. Taken together, these ndings
CLINICAL PHARMACOLOGY & THERAPEUTICS
146 Malaiyandi, Sellers, and Tyndale MARCH 2005
suggest that nicotine likely drives the urge to
smoke
51
and that this urge may be partly affected by
differing rates of nicotine metabolism. Genetic varia-
tion in CYP2A6, the main nicotine-inactivating en-
zyme, will be the focus of the remainder of this review.
INTERINDIVIDUAL AND INTERETHNIC
VARIATION IN NICOTINE METABOLISM
In humans approximately 70% to 80% of nicotine is
metabolized to cotinine
52
and roughly 90% of this
conversion is mediated by CYP2A6.
53,54
Cotinine is
subsequently oxidized, a step specically catalyzed by
CYP2A6 in vivo and in vitro, to form
3-hydroxycotinine.
55,56
CYP2A6 exhibits a narrow
substrate spectrum mediating the metabolism of cou-
marin, a few pharmaceutical compounds (eg, halo-
thane, valproic acid, disulram, and the antineoplastic
agent tegafur), and tobacco-specic nitrosamines.
57,58
Substantial variation in CYP2A6 activity and protein
levels exists. In vivo studies using coumarin
59-61
and
nicotine
52,55,62
as substrates indicate large interindi-
vidual variability in CYP2A6 activity. In addition, in
vitro kinetic studies using human liver microsomes
revealed interindividual variability in cotinine-to-
nicotine metabolic ratios.
53,54
Considerable interethnic
variability in coumarin
63
and nicotine
64
metabolism has
also been described. The presence of environmental or
therapeutic inducers and liver abnormality may contrib-
ute to these differences
58
; however, this variability has
been largely attributed to genetic polymorphisms in the
CYP2A6 gene.
65,66
Presently, there are 26 known variants of the
CYP2A6 gene that may affect enzyme activity or gene
regulation.
67
Table I summarizes the variants for which
functional consequences have been characterized in vitro or
in vivo. A number of alleles with single nucleotide polymor-
phisms in regulatory regions (CYP2A6*1B, CYP2A6*1C,
CYP2A6*1D, CYP2A6*1E, and CYP2A6*1H, CYP2A6*1J)
and coding regions (CYP2A6*1F, CYP2A6*1G, and
CYP2A6*13-*16) have also been identied; however,
little is known about their frequencies or effect on
function.
The frequencies of CYP2A6 alleles vary among
ethnicities (Table II). Generally, variant alleles re-
sulting in reduced enzyme activity are more common
among Asian populations (Chinese, Japanese, and
Korean) compared with Caucasian and African North
American populations (Fig 1).
64,68
In addition, cer-
Table I. Description of CYP2A6 alleles with known in vivo or in vitro functional consequences
CYP2A6
allele Nucleotide and structural changes Functional consequence Reference No.
*1A Normal activity
*8 6600GT (R485L) Normal in vivo activity 121, 122, 122a
*1x2 Reciprocal of *4A or *4D (gene duplication) Increased nicotine clearance in vivo 73, 74
*2 1799TA (L160H) No in vivo or in vitro activity 123, 124
*4A Unequal crossover (3-UTR) with CYP2A7
(gene deletion)
No in vivo activity 125-127
*4B Unequal crossover (distal 3-UTR) with
CYP2A7 (gene deletion)
No in vivo activity 128
*4D Unequal crossover (intron 8) with CYP2A7
(gene deletion)
No in vivo activity 74, 129
*5 6582GT (G479L) No in vivo or in vitro activity 129
*6 1703GA (R128Q) Reduced in vitro activity 130
*7 6558TC (I471T) Reduced in vivo and in vitro activity toward
nicotine but not coumarin
121, 122, 122a
*10 6558TC 6600GT (I471T R485L) Absent or reduced in vivo activity 122, 122a, 131
*11 3391TC (S224P) Reduced in vivo and in vitro activity toward
tegafur, an antineoplastic agent
132
*12 Hybrid gene CYP2A7: 5 regulatory to exon 2
CYP2A6: exon 3-3-UTR (10amino acid
substitution)
Reduced in vivo and in vitro activity toward
coumarin
133
*9 48TG (TATATAGA box) Decreased enzyme expression resulting in
reduced activity in vivo and in vitro
134-136
UTR, Untranslated region.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):145-58 CYP2A6 genetic variation and smoking 147
tain alleles vary substantially among Asian popula-
tions. For instance, the CYP2A6*4 allele frequency is
signicantly higher among Japanese subjects com-
pared with Chinese subjects.
68
Interethnic differ-
ences in CYP2A6-mediated cotinine formation
within genotype groups have also been described;
Korean subjects show substantially greater cotinine-
to-nicotine metabolic ratios compared with Japanese
subjects.
64
Although the frequency of variant alleles
is greater in Asian populations, smoking prevalence
is also much higher in this group.
69,70
However, even
among this population with a high smoking preva-
lence, the inuence of CYP2A6 on smoking is seen.
For example, Japanese poor metabolizers of CYP2A6
are less often found to be smokers, smoke less per
day if regular smokers, and have a reduced risk for
lung cancer.
71
However, this gene is only one part of
the genetic inuence, and in addition to genetic dif-
Fig 1. Frequencies reported as percentages by predicted activity. Calculations of genotype fre-
quencies are based on mean allelic frequencies and Hardy-Weinberg estimations of homozygotes
and heterozygotes. No CYP2A6 activity, Homozygous for CYP2A6*2, CYP2A6*4, and CYP2A6*5
inactive alleles; 50% CYP2A6 activity, heterozygous for inactive alleles and homozygous for any
combination of reduced activity alleles CYP2A6*6, CYP2A6*7, CYP2A6*9, CYP2A6*10, and
CYP2A6*12; 75% CYP2A6 activity, heterozygous for reduced activity alleles; Normal (100%)
CYP2A6 activity, remaining percentage of those not included in any other group. The African North
American ethnic group includes both African Americans and African Canadians.
Table II. Interethnic variation in CYP2A6 allele frequencies
CYP2A6
allele
Allele frequency
African North
American Caucasian Chinese Japanese Korean
*1x2 1.7%
74
0.5%
131
*2 0.3%-1.1%
68,137
1.1%-3.0%
68,74,124,135,138
0%-0.7%
68,127,138
0%
64,68
0%
64
*4 1.9%
68
0.5%-1.2%
68,74,127
6.7%-15.1%
68,127
20.1%-24.2%
64,68
11.0%
64
*5 0%
68
0%-0.1%
68,129
0%-0.5%
68,129
0%
64,68
0%-0.5%
64
*6 0%-0.4%
68,130,131
0%
131
*7 0%
68
0%-0.3%
68,122
2.2%-9.8%
68,122,122a
6.3%-12.5%
64,68,122a
3.6%
64
*8 0%
68
0%-0.1%
68,122
0%-3.6%
68,122,122a
0%-2.2%
64,68,122a
1.4%
64
*9 7.1%
68
5.2%-7.2%
68,135
15.6%-15.7%
68,135
20.3%-21.3%
68,136
22.3%
136
*10 0%
68
0%
68,122
0.4%-4.0%
68,122,122a
1.1%-3.2%
64,68,122a
0.5%
64
*12 0.4%
68
2.0%-2.2%
68,133
0%
68,133
0.8%
68
The African North American ethnic group includes African Americans and African Canadians.
CLINICAL PHARMACOLOGY & THERAPEUTICS
148 Malaiyandi, Sellers, and Tyndale MARCH 2005
ferences, environmental and cultural differences may
be inuential in driving smoking behavior. This may
be specic to populations such as the Chinese and
Japanese, in which smoking is socially and culturally
accepted and encouraged in men.
70
IMPACT OF CYP2A6 GENETIC VARIATION
ON NICOTINE DEPENDENCE AND SMOKING
BEHAVIORS
Susceptibility to nicotine dependence: Is slow me-
tabolism analogous to innate sensitivity? Several lines
of evidence indicate that smokers titrate their cigarette
consumption to maintain steady levels of nicotine in the
brain. It has been demonstrated that factors inuencing
nicotine plasma levels, by either intake (eg, high- ver-
sus low-yield cigarettes) or removal (eg, acidication
of urine to increase nicotine clearance), affect smoking
behaviors.
50,72
Variability in the rates of nicotine me-
tabolism, as a result of CYP2A6 genetic polymor-
phisms, alter nicotine plasma levels.
66,73,74
It was orig-
inally hypothesized by our laboratory that CYP2A6
genetic variation would affect smoking behaviors such
as the amount smoked and, secondarily, the risk for
nicotine dependence.
75
Specically, we hypothesized
that nicotine-dependent slow inactivators, individuals
with at least 1 inactive allele (CYP2A6*2 or
CYP2A6*4), would smoke fewer cigarettes and would
smoke less often (longer interval between cigarettes) to
avoid withdrawal symptoms compared with normal
inactivators.
76
In contrast, when learning to smoke,
slow nicotine inactivators might have prolonged or
elevated nicotine levels, which could alter the aversive
effects of nicotine. Consequently, we expected slow
inactivators to be at lower risk for dependence.
76
It is equally plausible that prolonged levels of brain
nicotine might increase the risk for dependence. The
sensitivity model proposes that those who have an
innate sensitivity to nicotine experience greater ef-
fects of nicotine, whether pleasurable or aversive, and
are more likely to progress to dependent smoking.
77,78
A retrospective study found that women who became
highly dependent smokers experienced greater pleasur-
able effects (rush, buzz, and relaxation) on initial
tobacco exposure compared with never-smokers.
79
These ndings are consistent with a prospective study
in youth showing that increased feelings of relaxation,
dizziness, and nausea after the rst nicotine exposure
may be risk factors for progression to nicotine depen-
dence.
80
Although individual pharmacokinetic effects
on nicotine levels and resulting nicotine sensitivity are
not well understood,
77
in this context one could think of
slow inactivators as being analogous to those who are
nicotine sensitive; they may experience greater nicotine
effects and thus may be more susceptible to nicotine
dependence.
CYP2A6 genetic variation and smoking: Retrospec-
tive and prospective studies. A number of studies in
adults have demonstrated that CYP2A6 genetic varia-
tion, causing reduced or absent enzyme activity, is
associated with a reduced risk for smoking, lower
amount smoked, altered smoking intensity, and in-
creased quitting
68,74,81-84
; however, not all studies are
in agreement with these ndings.
85-87
A meta-analysis
reviewing several studies on CYP2A6 and smoking
found no association between individuals with variant
alleles and smoking status or amount smoked.
88
As the
authors mention, the discordance in the literature may
be a result of various aspects of study design such as
broad denitions of smokers (ever-smoker, never-
smoker, former smoker, nonsmoker), population strat-
ication, genotype groups, and genotyping assays.
A recent study from our laboratory has addressed
some of these factors in an attempt to clarify the asso-
ciation between CYP2A6 genetic variation and smok-
ing.
68
Substantial differences in allele frequencies were
observed among ethnic groups. To minimize popula-
tion stratication, the smoking study population was
restricted to adults of Caucasian ancestry determined by
participants reports of having at least 3 grandparents of
Caucasian ethnicity. In this study the effect of the gene
on smoking maintenance, not initiation, was investi-
gated. To assess the effect of the gene (CYP2A6) on the
behavior (smoking), at least a single exposure to the
drug (nicotine) is necessary. Therefore in this study the
nonsmoker group was restricted to individuals who had
never become current smokers but had smoked 1 to 99
cigarettes in their lifetime. Smokers, those currently
smoking (100 cigarettes/lifetime), were stratied by
dependence. Those who met criteria for tobacco depen-
dence according to the American Psychiatric Associa-
tion Diagnostic and Statistical Manual of Mental Dis-
orders, Fourth Edition, were dened as dependent
smokers. Only variant allele products known to be
inactive or to have decreased activity with frequencies
greater than 1% in Caucasians were studied. Genotype
groups were segregated on the basis of combined the-
oretic and demonstrated CYP2A6 activity as slow in-
activators (50% activity), intermediate inactivators
(approximately 75% activity), and normal inactivators
(full activity).
Applying these specications to the study design and
data analysis, we found that several aspects of smoking
behavior were inuenced by CYP2A6 genotype. Slow
inactivators were signicantly more prevalent among
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):145-58 CYP2A6 genetic variation and smoking 149
nonsmokers than smokers independent of tobacco de-
pendence (odds ratio, 0.52 [95% condence interval
(CI), 0.30-0.91]; P .027). This indicates that CYP2A6
slow inactivators of nicotine are at lower risk for being
current smokers regardless of whether they are
nicotine-dependent. CYP2A6 also affected the amount
smoked per day. Nicotine-dependent slow inactivators
smoked fewer cigarettes per day compared with normal
inactivators (21.3 cigarettes per day versus 28.2 ciga-
rettes per day, P .003), but this gene effect was not
found in individuals who were current nondependent
smokers. This highlights the importance of assessing
dependence because the gene effect on the amount
smoked may not be detected in groups in which there is
a greater proportion of nondependent smokers.
68
This
nding is consistent with our original hypothesis on the
amount smoked.
74,75
We expected that the titration of
plasma and brain nicotine levels by cigarette number
would occur only in dependent smokers.
49,76
Are slow nicotine-inactivator genotypes less preva-
lent among smokers because of a lower risk of becom-
ing a smoker or because these individuals quit smoking
sooner? In our adult study a trend toward a shorter
duration of smoking was observed among current
smokers who were slow inactivators. We found that the
proportion of slow inactivators was greatest among
nonsmokers, followed by short-duration smokers (10
years), and was lowest among long-duration smokers
(30 years), suggesting that smokers with slow inac-
tivator genotypes may be quitting sooner. This inter-
pretation is supported by studies assessing the effect of
CYP2A6 genotype on quitting. A smoking cessation
study conducted in African Americans showed that
slow inactivators had substantially increased cessation
rates (odds ratio, 3.8 [95% CI, 1.4-11.1]) compared
with normal inactivators.
89
In addition, Caucasians
having a CYP2A6*2 allele were 1.75 (95% CI, 1.17-
2.61) times more likely to quit compared with those
without the CYP2A6*2 allele.
90
These studies provide
support that smokers with a slow inactivator genotype
may be quitting sooner and, consequently, are under-
represented in smoker groups, particularly as the dura-
tion of smoking increases. However, there are still slow
inactivators who remain among long-duration smokers;
thus it will be important to determine whether these
slow inactivators continue smoking because they are
resistant to quitting or whether they have not yet made
serious attempts to stop smoking. A recent assessment
in a nicotine replacement clinical trial indicates that a
very low percentage of slow metabolizers enrolled (Ler-
man C, Malaiyandi V, Tyndale RF, unpublished data,
September 2004); this suggests that slow metabolizers
may be quitting before this time or may not be seeking
treatment. The apparent effect of CYP2A6 on increas-
ing quitting, particularly among shorter-duration smok-
ers, may contribute to the lack of agreement in pub-
lished studies between CYP2A6 slow inactivators and
decreased risk for smoking. A signicantly greater
prevalence of slow inactivator compared with normal
inactivator genotypes may not be apparent if the study
population is composed of smokers who have only been
smoking for a short duration (eg, 0-15 years).
To further investigate the role of CYP2A6 in smoking
and nicotine dependence, we examined the effect of
genetic variation on becoming a smoker. Retrospective
studies are adequate for examining the risk for current
or former smoking. However, they are weaker for as-
sessing the effect of CYP2A6 genetic variation on
smoking initiation and the risk for development of
tobacco dependence. Adults may have difculty accu-
rately recalling all aspects of their smoking histories,
especially during initiation, such as their experiences of
nicotines effects when rst learning to smoke. A pro-
spective study conducted in Caucasian adolescents fur-
ther elucidates the role of CYP2A6 genetic variation on
the risk for acquisition of nicotine dependence.
91
This
study followed an adolescent cohort to assess whether
CYP2A6 genetic variants altered the risk for becoming
nicotine-dependent from the point of rst inhalation.
The incidence of conversion to tobacco dependence
was approximately 3 times greater (hazard ratio, 2.8
[95% CI, 1.3-6.3]) among slow inactivators.
91
Although this nding is inconsistent with our origi-
nal hypothesis,
76
it is consistent with our ndings in
adults, in which a younger age of rst smoking was
reported among slow inactivators compared with nor-
mal inactivators (13 years versus 14.2 years, P
.03).
68
With poor recall accounted for, this result may
actually represent the age at which regular smoking was
established and appears to be consistent with a greater
occurrence of becoming tobacco-dependent among ad-
olescent slow inactivators.
91
Furthermore, in the ado-
lescent study it was observed that the prevalence of
smoking among parents of slow inactivators was lower
compared with parents of normal inactivators. The re-
duced frequency of smoking among parents of slow
inactivators, the presumed source of slow inactivator
alleles, is consistent with the reduced frequency of
smoking observed among adult slow inactivators. This
observation is also consistent with the reduced fre-
quency of slow inactivators among smokers as the
duration of smoking increases.
68
An increased incidence of conversion to tobacco
dependence among adolescent slow inactivators after
CLINICAL PHARMACOLOGY & THERAPEUTICS
150 Malaiyandi, Sellers, and Tyndale MARCH 2005
rst inhalation may also be consistent with the sensi-
tivity model. Slow rates of nicotine metabolism, pre-
dicted by CYP2A6 genotype, may cause more intense
and prolonged exposure of the brain to nicotine, which
may be analogous to increased nicotine sensitivity,
resulting in greater vulnerability to dependence. Ado-
lescent slow inactivators reported no differences in
symptoms of dizziness or nausea compared with nor-
mal inactivators.
91
It appears that when adolescents are
learning to smoke, slow nicotine inactivation, rather
than being protective against nicotine dependence by
enhancing the drugs aversive effects, may increase
susceptibility to dependence by greatly enhancing nic-
otines reinforcing effects. The sensitivity model sug-
gests that increased reinforcement, but not necessarily
increased aversive effects of nicotine, is predictive of
dependent smoking.
77-79
Although adolescent slow inactivators are becoming
dependent on nicotine sooner, cigarette consumption
among dependent adolescent smokers is markedly
lower in slow inactivators compared with normal inac-
tivators.
91
This nding is consistent with several studies
in adults
68,74,82
and supports the proposal that the gene
effect on the amount smoked is likely occurring only
among dependent smokers, whether in adults or ado-
lescents. In addition, the amount smoked by dependent
intermediate inactivators fell between the amounts of
slow and normal inactivators.
91
These adolescent data
parallel those of our adult study implicating a gene-
dose effect among dependent smokers and imply that
relatively small changes in nicotine metabolism may
affect the amount smoked.
68
Also, in support of the
gene effect on cigarette consumption, a kinetic study in
adults found that the metabolic ratio of
3-hydroxycotinine to cotinine, representing CYP2A6
activity, was signicantly correlated with the number of
cigarettes smoked per day, suggesting that the rate of
CYP2A6-mediated nicotine inactivation is a determi-
nant of the amount smoked.
92
Moreover, the adolescent
study suggests that among novice smokers slow nico-
tine inactivators are at greater risk for nicotine depen-
dence even though they are smoking less than normal
inactivators.
91
Temporal effect of CYP2A6 genetic variation:
Stages of smoking? Several studies in animals and
humans suggest that the adolescent brain is much more
sensitive to nicotine than the adult brain.
80,93-97
It has
been shown that adolescents show symptoms of nico-
tine dependence even if smoking 1 cigarette per
month.
96
Studies in the adolescent rat demonstrated that
even brief and low exposure to nicotine, similar to that
seen in adolescent smokers, results in the up-regulation
of nicotinic acetylcholine receptors in brain regions
associated with nicotine dependence, long-term reduc-
tions in cholinergic synaptic activity,
93
and increased
intravenous nicotine self-administration.
94
Taken to-
gether, these studies suggest that the adolescent brain
may be more susceptible to nicotine dependence as a
result of increased susceptibility to nicotines molecular
and behavioral effects. It is interesting that the potential
age-dependent effect of the CYP2A6 gene is consistent
with the period during which the brain appears to be
most sensitive to nicotines neuroplastic and behavioral
effects. It could be postulated that prolonged elevated
brain nicotine levels, as a result of CYP2A6 slow inac-
tivator alleles, may make adolescent slow inactivators
more vulnerable to nicotine dependence compared with
normal inactivators.
Collectively, these recent studies suggest that several
aspects of smoking are inuenced by CYP2A6 genetic
variation.
68,89,91
More specically, these effects appear
to vary with the point in smoking history and are
inuenced by factors such as status of nicotine depen-
dence and duration of smoking (Fig 2). Our studies
suggest that, during adolescence, having a slow inacti-
vator genotype increases the risk for nicotine depen-
dence while still reducing the amount smoked. As these
dependent novice smokers become adults and more
regular smoking occurs, it appears that slow nicotine
inactivators smoke less and for fewer years compared
with normal nicotine inactivators. Reduced cigarette
consumption has been associated with better cessation
outcomes in adults
98,99
; thus CYP2A6-mediated slow
nicotine inactivation may alter tolerance and with-
drawal such that these individuals are able to quit
sooner.
45
To further understand the temporal effect of
CYP2A6 genetic variation on smoking, we are continu-
ing to follow up this group of adolescents into adult-
hood. In addition, we are conducting a prospective
smoking study in college students in which we will be
able to assess the effect of CYP2A6 on the acquisition
of nicotine dependence and smoking behaviors in
young adults.
CYP2A6 GENETIC VARIATION:
IMPLICATIONS FOR SMOKING CESSATION
AND RISK FOR TOBACCO-RELATED
CANCERS
Cigarette smoking was the cause of 4.83 million
premature deaths globally in the year 2000.
100
Quitting
smoking has both immediate and long-term health ben-
ets such as reducing the risk for cardiovascular dis-
eases
101
and cancer.
102
Currently, nicotine replacement
therapy, including transdermal patch, nasal spray, gum
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):145-58 CYP2A6 genetic variation and smoking 151
inhaler, and lozenges, is the most commonly used treat-
ment for smoking cessation.
103
Although these treat-
ments double the odds for quitting,
104
80% to 90% of
those who try these treatments have a relapse to their
former smoking patterns.
105
In addition to nicotine, 2 tobacco-specic nitro-
samines, 4-(methylnitrosamino)-1-(3-pyridyl)-1-
butanone (NNK) and N-nitrosonornicotine (NNN), are
metabolized by CYP2A6 to form active lung
carcinogens.
106-109
It has been shown that the risk for
cancer is related to the amount of smoke exposure,
110
and studies have found that the CYP2A6*4 inactive
allele is associated with a lower risk of lung can-
cer,
82,111,112
but not all studies are in agreement with
this nding.
86,113,114
Taken together, these results sug-
gest that smokers with CYP2A6 slow inactivator geno-
types may be at a lower risk for lung cancer as a result
of reduced bioactivation of these procarcinogens. In
addition, these individuals have lower cigarette con-
sumption and smoke for fewer years,
68
which will
reduce their overall exposure to these procarcino-
gens.
71,103
As mentioned, dependent smokers regulate their
smoking to maintain target nicotine concentrations in
the brain,
49,68
and CYP2A6 genetic variation affects
systemic levels of nicotine.
66,73
Inhibition of the
CYP2A6 enzyme would slow nicotine inactivation,
would prolong target brain nicotine levels, and would
be expected to reduce smoking behavior.
115
We have
done this phenocopying maneuver, using CYP2A6
inhibitors in normal metabolizers to mimic slow me-
tabolism, to demonstrate that the effect seen with ge-
netically reduced metabolism can also be reproduced
with inhibition of metabolism. In addition, CYP2A6
inhibition presents a novel treatment strategy. Methox-
salen is a potent CYP2A6 inhibitor currently used to
treat psoriasis.
116,117
Compared with placebo or nicoti-
nine alone, methoxsalen can signicantly increase
plasma nicotine levels. In addition, it reduced subjects
self-reported desire to smoke, decreased the amount
smoked, increased the latency to the next cigarette, and
decreased the total number of puffs taken.
118
In addi-
tion to altering smoking, inhibition of CYP2A6 is ex-
pected to decrease activation of procarcinogen sub-
strates. Methoxsalen, given to dependent smokers for 3
days during smoking as desired, also signicantly re-
duced smoke exposure and increased the formation of
the inactive metabolite of NNK, suggesting rerouting
from the procarcinogen activating pathway.
115,119
To-
gether, genetic and biochemical studies indicate that
inhibiting CYP2A6 in smokers may reduce the amount
smoked, reduce the activation of tobacco-specic pro-
carcinogens, and increase quitting.
115,120
Fig 2. CYP2A6 slow nicotine inactivators. Adolescent slow nicotine inactivators experimenting
with smoking may experience prolonged nicotine levels in the brain, which may increase their risk
for becoming nicotine-dependent, although their cigarette consumption, once nicotine-dependent, is
lower. As smoking continues into adulthood, the amount smoked daily remains lower and quitting
may increase for slow inactivators. The reduced cigarette consumption throughout a slow inacti-
vators smoking history may contribute to the lower risk for being a smoker, via increased quitting,
as the duration of smoking increases.
CLINICAL PHARMACOLOGY & THERAPEUTICS
152 Malaiyandi, Sellers, and Tyndale MARCH 2005
CONCLUSIONS
In summary, we conclude that slow nicotine inacti-
vation, predicted by CYP2A6 genetic variation, inu-
ences the risk for several aspects of smoking behavior
and nicotine dependence. However, the inuence of
CYP2A6 appears to vary along the smoking continuum
and is affected by factors such as status of nicotine
dependence and duration of smoking. The slow inacti-
vator genotype may increase the risk for nicotine de-
pendence when smoking is initiated during adolescence
but reduces the risk for smoking, lowers the amount
smoked, and decreases the duration of smoking among
adult dependent smokers. The apparent temporal effect
of CYP2A6 on smoking behavior and nicotine depen-
dence requires further investigation, but we postulate
that the continued effect of slow metabolism on reduc-
ing cigarette consumption, throughout the smoking his-
tory of slow inactivators, may affect tolerance and
withdrawal mechanisms, facilitating quitting among
these individuals. These ndings have implications for
the use of existing smoking cessation treatments in
adolescents and support CYP2A6 phenocopying as a
novel treatment for smoking reduction and cessation.
Moreover, the potential temporal effect of CYP2A6
genetic variation emphasizes the importance of un-
derstanding the various factors modulating vulnerabil-
ity to nicotine dependence and the need to identify
more specic smoking and disease prevention strate-
gies.
121,123,125,126,128,132,134
R.F.T. and E.M.S. hold shares in Nicogen Inc (Toronto, Ontario,
Canada) and are involved in the research program undertaken by
Nicogen Inc, a research pharmaceutical company that is focused on
the development of novel smoking cessation treatments based on the
inhibition of nicotine metabolism. V.M. is a student in the laboratory
of Dr Tyndale and has no involvement with Nicogen Inc.
References
1. Henningeld JE, Miyasato K, Jasinski DR. Abuse lia-
bility and pharmacodynamic characteristics of intrave-
nous and inhaled nicotine. J Pharmacol Exp Ther 1985;
234:1-12.
2. Stolerman IP, Jarvis MJ. The scientic case that nico-
tine is addictive. Psychopharmacology (Berl) 1995;117:
2-10, discussion 14-20.
3. Pidoplichko VI, DeBiasi M, Williams JT, Dani JA.
Nicotine activates and desensitizes midbrain dopamine
neurons. Nature 1997;390:401-4.
4. Pontieri FE, Tanda G, Orzi F, Di Chiara G. Effects of
nicotine on the nucleus accumbens and similarity to
those of addictive drugs. Nature 1996;382:255-7.
5. Salokangas RK, Vilkman H, Ilonen T, Taiminen T,
Bergman J, Haaparanta M, et al. High levels of dopa-
mine activity in the basal ganglia of cigarette smokers.
Am J Psychiatry 2000;157:632-4.
6. Laviolette SR, van der Kooy D. Blockade of mesolim-
bic dopamine transmission dramatically increases sen-
sitivity to the rewarding effects of nicotine in the ventral
tegmental area. Mol Psychiatry 2003;8:50-9.
7. Stein EA, Pankiewicz J, Harsch HH, Cho JK, Fuller SA,
Hoffmann RG, et al. Nicotine-induced limbic cortical
activation in the human brain: a functional MRI study.
Am J Psychiatry 1998;155:1009-15.
8. Koob GF. Neurobiology of addiction. Toward the de-
velopment of new therapies. Ann N Y Acad Sci 2000;
909:170-85.
9. Baker TB, Brandon TH, Chassin L. Motivational inu-
ences on cigarette smoking. Annu Rev Psychol 2004;
55:463-91.
10. Buisson B, Bertrand D. Nicotine addiction: the possible
role of functional upregulation. Trends Pharmacol Sci
2002;23:130-6.
11. Flores CM, Davila-Garcia MI, Ulrich YM, Kellar KJ.
Differential regulation of neuronal nicotinic receptor
binding sites following chronic nicotine administration.
J Neurochem 1997;69:2216-9.
12. Perry DC, Davila-Garcia MI, Stockmeier CA, Kellar
KJ. Increased nicotinic receptors in brains from smok-
ers: membrane binding and autoradiography studies.
J Pharmacol Exp Ther 1999;289:1545-52.
13. Epping-Jordan MP, Watkins SS, Koob GF, Markou A.
Dramatic decreases in brain reward function during
nicotine withdrawal. Nature 1998;393:76-9.
14. Teneggi V, Tiffany ST, Squassante L, Milleri S, Ziviani
L, Bye A. Smokers deprived of cigarettes for 72 h:
effect of nicotine patches on craving and withdrawal.
Psychopharmacology (Berl) 2002;164:177-87.
15. Benowitz NL. Pharmacology of nicotine: addiction and
therapeutics. Annu Rev Pharmacol Toxicol 1996;36:
597-613.
16. Jarvis MJ. Why people smoke. BMJ 2004;328:277-9.
17. Picciotto MR. Nicotine as a modulator of behavior:
beyond the inverted U. Trends Pharmacol Sci 2003;24:
493-9.
18. Mayhew KP, Flay BR, Mott JA. Stages in the develop-
ment of adolescent smoking. Drug Alcohol Depend
2000;59(Suppl 1):S61-81.
19. Lerman C, Berrettini W. Elucidating the role of genetic
factors in smoking behavior and nicotine dependence.
Am J Med Genet 2003;118B:48-54.
20. Sullivan PF, Kendler KS. The genetic epidemiology of
smoking. Nicotine Tob Res 1999;1(Suppl 2):S51-7, dis-
cussion S69-70.
21. Tyndale RF. Genetics of alcohol and tobacco use in
humans. Ann Med 2003;35:94-121.
22. Heath AC, Kirk KM, Meyer JM, Martin NG. Genetic
and social determinants of initiation and age at onset of
smoking in Australian twins. Behav Genet 1999;29:
395-407.
23. Koopmans JR, Slutske WS, Heath AC, Neale MC,
Boomsma DI. The genetics of smoking initiation and
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):145-58 CYP2A6 genetic variation and smoking 153
quantity smoked in Dutch adolescent and young adult
twins. Behav Genet 1999;29:383-93.
24. Madden PA, Heath AC, Pedersen NL, Kaprio J, Kosk-
envuo MJ, Martin NG. The genetics of smoking persis-
tence in men and women: a multicultural study. Behav
Genet 1999;29:423-31.
25. Lessov CN, Benowitz NL, Swan GE. Evidence for
shared genetic risk between depressive symptoms and
smoking experimentation in men [abstract]. In Proceed-
ings of the Tenth Annual Meeting of the Society for
Research on Nicotine and Tobacco; 2004 Feb 18-21;
Scottsdale, Arizona. Scottsdale (AZ): The Society;
2004.
26. De Luca V, Wong AH, Muller DJ, Wong GW, Tyndale
RF, Kennedy JL. Evidence of association between
smoking and alpha7 nicotinic receptor subunit gene in
schizophrenia patients. Neuropsychopharmacology
2004;29:1522-6.
27. Freedman R, Adler LE, Bickford P, Byerley W, Coon
H, Cullum CM, et al. Schizophrenia and nicotinic re-
ceptors. Harv Rev Psychiatry 1994;2:179-92.
28. Adler LE, Hoffer LJ, Grifth J, Waldo MC, Freedman
R. Normalization by nicotine of decient auditory sen-
sory gating in the relatives of schizophrenics. Biol Psy-
chiatry 1992;32:607-16.
29. Freedman R, Coon H, Myles-Worsley M, Orr-Urtreger
A, Olincy A, Davis A, et al. Linkage of a neurophysi-
ological decit in schizophrenia to a chromosome 15
locus. Proc Natl Acad Sci U S A 1997;94:587-92.
30. Uhl GR, Liu QR, Walther D, Hess J, Naiman D. Poly-
substance abuse-vulnerability genes: genome scans for
association, using 1,004 subjects and 1,494 single-
nucleotide polymorphisms. Am J Hum Genet 2001;69:
1290-300.
31. Li MD, Ma JZ, Cheng R, Dupont RT, Williams NJ,
Crews KM, et al. A genome-wide scan to identify loci
for smoking rate in the Framingham Heart Study pop-
ulation. BMC Genet 2003;4(Suppl 1):S103-7.
32. Sullivan PF, Neale BM, van den Oord E, Miles MF,
Neale MC, Bulik CM, et al. Candidate genes for nico-
tine dependence via linkage, epistasis, and bioinformat-
ics. Am J Med Genet 2004;126B:23-36.
33. Wilhelmsen KC, Feiler HS, S-Cheng L, Lessov CN,
Ring HZ, Jack LM, et al. A genome-wide search for
quantitative trait loci associated with tobacco use phe-
notypes [abstract]. In Proceedings of the Tenth Annual
Meeting of the Society for Research on Nicotine and
Tobacco; 2004 Feb 18-21; Scottsdale, Arizona. Scotts-
dale (AZ): The Society; 2004.
34. Comings DE, Blum K. Reward deciency syndrome:
genetic aspects of behavioral disorders. Prog Brain Res
2000;126:325-41.
35. Goudriaan AE, Oosterlaan J, de Beurs E, Van den Brink
W. Pathological gambling: a comprehensive review of
biobehavioral ndings. Neurosci Biobehav Rev 2004;
28:123-41.
36. Arinami T, Ishiguro H, Onaivi ES. Polymorphisms in
genes involved in neurotransmission in relation to
smoking. Eur J Pharmacol 2000;410:215-26.
37. Batra V, Patkar AA, Berrettini WH, Weinstein SP,
Leone FT. The genetic determinants of smoking. Chest
2003;123:1730-9.
38. Lerman C, Niaura R. Applying genetic approaches to
the treatment of nicotine dependence. Oncogene 2002;
21:7412-20.
39. Benowitz NL, Zevin S, Jacob P III. Suppression of
nicotine intake during ad libitum cigarette smoking by
high-dose transdermal nicotine. J Pharmacol Exp Ther
1998;287:958-62.
40. Rose JE, Corrigall WA. Nicotine self-administration in
animals and humans: similarities and differences. Psy-
chopharmacology (Berl) 1997;130:28-40.
41. Klink R, de Kerchove dExaerde A, Zoli M, Changeux
JP. Molecular and physiological diversity of nicotinic
acetylcholine receptors in the midbrain dopaminergic
nuclei. J Neurosci 2001;21:1452-63.
42. Nguyen HN, Rasmussen BA, Perry DC. Subtype-
selective up-regulation by chronic nicotine of high-
afnity nicotinic receptors in rat brain demonstrated by
receptor autoradiography. J Pharmacol Exp Ther 2003;
307:1090-7.
43. Picciotto MR, Zoli M, Rimondini R, Lena C, Marubio
LM, Pich EM, et al. Acetylcholine receptors containing
the beta2 subunit are involved in the reinforcing prop-
erties of nicotine. Nature 1998;391:173-7.
44. Marubio LM, Gardier AM, Durier S, David D, Klink R,
Arroyo-Jimenez MM, et al. Effects of nicotine in the
dopaminergic system of mice lacking the alpha4 subunit
of neuronal nicotinic acetylcholine receptors. Eur
J Neurosci 2003;17:1329-37.
45. Schilstrom B, Svensson HM, Svensson TH, Nomikos
GG. Nicotine and food induced dopamine release in the
nucleus accumbens of the rat: putative role of alpha7
nicotinic receptors in the ventral tegmental area. Neu-
roscience 1998;85:1005-9.
46. Nomikos GG, Hildebrand BE, Panagis G, Svensson TH.
Nicotine withdrawal in the rat: role of alpha7 nicotinic
receptors in the ventral tegmental area. Neuroreport
1999;10:697-702.
47. Feng Y, Niu T, Xing H, Xu X, Chen C, Peng S, et al. A
common haplotype of the nicotine acetylcholine recep-
tor alpha 4 subunit gene is associated with vulnerability
to nicotine addiction in men. Am J Hum Genet 2004;
75:112-21.
48. Lueders KK, Hu S, McHugh L, Myakishev MV, Sirota
LA, Hamer DH. Genetic and functional analysis of
single nucleotide polymorphisms in the beta2-neuronal
nicotinic acetylcholine receptor gene (CHRNB2). Nic-
otine Tob Res 2002;4:115-25.
49. McMorrow MJ, Foxx RM. Nicotines role in smoking:
an analysis of nicotine regulation. Psychol Bull 1983;
93:302-27.
CLINICAL PHARMACOLOGY & THERAPEUTICS
154 Malaiyandi, Sellers, and Tyndale MARCH 2005
50. Benowitz NL, Jacob P III. Nicotine renal excretion rate
inuences nicotine intake during cigarette smoking.
J Pharmacol Exp Ther 1985;234:153-5.
51. Jarvik ME, Madsen DC, Olmstead RE, Iwamoto-
Schaap PN, Elins JL, Benowitz NL. Nicotine blood
levels and subjective craving for cigarettes. Pharmacol
Biochem Behav 2000;66:553-8.
52. Benowitz NL, Jacob P III. Metabolism of nicotine to
cotinine studied by a dual stable isotope method. Clin
Pharmacol Ther 1994;56:483-93.
53. Messina ES, Tyndale RF, Sellers EM. A major role for
CYP2A6 in nicotine C-oxidation by human liver micro-
somes. J Pharmacol Exp Ther 1997;282:1608-14.
54. Nakajima M, Yamamoto T, Nunoya K, Yokoi T, Na-
gashima K, Inoue K, et al. Role of human cytochrome
P4502A6 in C-oxidation of nicotine. Drug Metab Dis-
pos 1996;24:1212-7.
55. Dempsey D, Tutka P, Jacob P III, Allen F, Schoedel K,
Tyndale RF, et al. Nicotine metabolite ratio as an index
of cytochrome P450 2A6 metabolic activity. Clin Phar-
macol Ther 2004;76:64-72.
56. Nakajima M, Yamamoto T, Nunoya K, Yokoi T, Na-
gashima K, Inoue K, et al. Characterization of CYP2A6
involved in 3-hydroxylation of cotinine in human liver
microsomes. J Pharmacol Exp Ther 1996;277:1010-5.
57. Pelkonen O, Rautio A, Raunio H, Pasanen M. CYP2A6:
a human coumarin 7-hydroxylase. Toxicology 2000;
144:139-47.
58. Xu C, Goodz S, Sellers EM, Tyndale RF. CYP2A6
genetic variation and potential consequences. Adv Drug
Deliv Rev 2002;54:1245-56.
59. Hadidi H, Irshaid Y, Vagbo CB, Brunsvik A, Cholerton
S, Zahlsen K, et al. Variability of coumarin 7- and
3-hydroxylation in a Jordanian population is suggestive
of a functional polymorphism in cytochrome P450
CYP2A6. Eur J Clin Pharmacol 1998;54:437-41.
60. Iscan M, Rostami H, Iscan M, Guray T, Pelkonen O,
Rautio A. Interindividual variability of coumarin
7-hydroxylation in a Turkish population. Eur J Clin
Pharmacol 1994;47:315-8.
61. Rautio A, Kraul H, Kojo A, Salmela E, Pelkonen O.
Interindividual variability of coumarin 7-hydroxylation
in healthy volunteers. Pharmacogenetics 1992;2:227-
33.
62. Benowitz NL, Jacob P III. Trans-3-hydroxycotinine:
disposition kinetics, effects and plasma levels during
cigarette smoking. Br J Clin Pharmacol 2001;51:53-9.
63. Shimada T, Yamazaki H, Guengerich FP. Ethnic-related
differences in coumarin 7-hydroxylation activities cat-
alyzed by cytochrome P4502A6 in liver microsomes of
Japanese and Caucasian populations. Xenobiotica 1996;
26:395-403.
64. Kwon JT, Nakajima M, Chai S, Yom YK, Kim HK,
Yamazaki H, et al. Nicotine metabolism and CYP2A6
allele frequencies in Koreans. Pharmacogenetics 2001;
11:317-23.
65. Nakajima M, Kwon JT, Tanaka N, Zenta T, Yamamoto
Y, Yamamoto H, et al. Relationship between interindi-
vidual differences in nicotine metabolism and CYP2A6
genetic polymorphism in humans. Clin Pharmacol Ther
2001;69:72-8.
66. Nakajima M, Yamagishi S, Yamamoto H, Yamamoto
T, Kuroiwa Y, Yokoi T. Decient cotinine formation
from nicotine is attributed to the whole deletion of the
CYP2A6 gene in humans. Clin Pharmacol Ther 2000;
67:57-69.
67. CYP2A6 allele nomenclature. Available from: URL:
http://www.imm.ki.se/CYPalleles/cyp2a6.htm. Ac-
cessed July 20, 2004.
68. Schoedel K, Hoffmann E, Rao Y, Sellers E, Tyndale
RF. Ethnic variation in CYP2A6 and association of
genetically slow nicotine metabolism and smoking in
adult Caucasians. Pharmacogenetics 2004;14:615-26.
69. Hsia FN, Spruijt-Metz D. The meanings of smoking
among Chinese American and Taiwanese American col-
lege students. Nicotine Tob Res 2003;5:837-50.
70. Spruijt-Metz D, Gallaher PE, Unger JB, Anderson-
Johnson C. Meanings of smoking and adolescent smok-
ing across ethnicities. J Adolesc Health 2004;35:197-
205.
71. Fujieda M, Yamazaki H, Saito T, Kiyotani K, Gyam
MA, Sakurai M, et al. Evaluation of CYP2A6 genetic
polymorphisms as determinants of smoking behavior
and tobacco-related lung cancer risk in male Japanese
smokers. Carcinogenesis 2004;25:2451-8.
72. Zacny JP, Stitzer ML. Cigarette brand-switching: ef-
fects on smoke exposure and smoking behavior. J Phar-
macol Exp Ther 1988;246:619-27.
73. Benowitz NL, Tyndale RF, Jacob P III, Swan GE.
CYP2A6 polymorphisms and nicotine metabolism [ab-
stract]. Clin Pharmacol Ther 2002;71:P41.
74. Rao Y, Hoffmann E, Zia M, Bodin L, Zeman M, Sellers
EM, et al. Duplications and defects in the CYP2A6
gene: identication, genotyping, and in vivo effects on
smoking. Mol Pharmacol 2000;58:747-55.
75. Pianezza ML, Sellers EM, Tyndale RF. Nicotine me-
tabolism defect reduces smoking. Nature 1998;393:750.
76. Tyndale RF, Sellers EM. Variable CYP2A6-mediated
nicotine metabolism alters smoking behavior and risk.
Drug Metab Dispos 2001;29:548-52.
77. Pomerleau OF. Individual differences in sensitivity to
nicotine: implications for genetic research on nicotine
dependence. Behav Genet 1995;25:161-77.
78. Pomerleau OF, Collins AC, Shiffman S, Pomerleau CS.
Why some people smoke and others do not: new per-
spectives. J Consult Clin Psychol 1993;61:723-31.
79. Pomerleau OF, Pomerleau CS, Namenek RJ. Early ex-
periences with tobacco among women smokers, ex-
smokers, and never-smokers. Addiction 1998;93:595-9.
80. DiFranza JR, Savageau JA, Fletcher K, Ockene JK,
Rigotti NA, McNeill AD, et al. Recollections and re-
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):145-58 CYP2A6 genetic variation and smoking 155
percussions of the rst inhaled cigarette. Addict Behav
2004;29:261-72.
81. Ando M, Hamajima N, Ariyoshi N, Kamataki T, Mat-
suo K, Ohno Y. Association of CYP2A6 gene deletion
with cigarette smoking status in Japanese adults. J Epi-
demiol 2003;13:176-81.
82. Ariyoshi N, Miyamoto M, Umetsu Y, Kunitoh H,
Dosaka-Akita H, Sawamura Y, et al. Genetic polymor-
phism of CYP2A6 gene and tobacco-induced lung can-
cer risk in male smokers. Cancer Epidemiol Biomarkers
Prev 2002;11:890-4.
83. Iwahashi K, Waga C, Takimoto T. Whole deletion of
CYP2A6 gene (CYP2A6*4C) and smoking behavior.
Neuropsychobiology 2004;49:101-4.
84. Minematsu N, Nakamura H, Iwata M, Tateno H, Naka-
jima T, Takahashi S, et al. Association of CYP2A6
deletion polymorphism with smoking habit and devel-
opment of pulmonary emphysema. Thorax 2003;58:
623-8.
85. London SJ, Idle JR, Daly AK, Coetzee GA. Genetic
variation of CYP2A6, smoking, and risk of cancer.
Lancet 1999;353:898-9.
86. Loriot MA, Rebuissou S, Oscarson M, Cenee S, Miy-
amoto M, Ariyoshi N, et al. Genetic polymorphisms of
cytochrome P450 2A6 in a case-control study on lung
cancer in a French population. Pharmacogenetics 2001;
11:39-44.
87. Sabol SZ, Hamer DH. An improved assay shows no
association between the CYP2A6 gene and cigarette
smoking behavior. Behav Genet 1999;157:632-4.
88. Carter B, Long T, Cinciripini P. A meta-analytic review
of the CYP2A6 genotype and smoking behavior. Nico-
tine Tob Res 2004;6:221-7.
89. Goodz S, Alhuwalia J, Harris K, Xu C, Rao YS, Xu B,
et al. CYP2A6 genetic variants among African-
Americans. In Proceedings of the Eighth Annual Meet-
ing of the Society for Research on Nicotine and To-
bacco; 2002 Feb 20-23; Savannah, Georgia. Savannah
(GA): The Society; 2002.
90. Gu DF, Hinks LJ, Morton NE, Day IN. The use of long
PCR to conrm three common alleles at the CYP2A6
locus and the relationship between genotype and smok-
ing habit. Ann Hum Genet 2000;64:383-90.
91. OLaughlin J, Paradis G, Kim W, DiFranza JR,
Meshefedijan G, McMillan E, et al. Genetically de-
creased CYP2A6 and the risk of tobacco dependence: a
prospective study in novice smokers. Tob Control 2004;
13:422-8.
92. Benowitz NL, Pomerleau OF, Pomerleau CS, Jacob P
III. Nicotine metabolite ratio as a predictor of cigarette
consumption. Nicotine Tob Res 2003;5:621-4.
93. Abreu-Villaca Y, Seidler FJ, Qiao D, Tate CA, Cousins
MM, Thillai I, et al. Short-term adolescent nicotine
exposure has immediate and persistent effects on cho-
linergic systems: critical periods, patterns of exposure,
dose thresholds. Neuropsychopharmacology 2003;28:
1935-49.
94. Adriani W, Spijker S, Deroche-Gamonet V, Laviola G,
Le Moal M, Smit AB, et al. Evidence for enhanced
neurobehavioral vulnerability to nicotine during peri-
adolescence in rats. J Neurosci 2003;23:4712-6.
95. Belluzzi JD, Lee AG, Oliff HS, Leslie FM. Age-
dependent effects of nicotine on locomotor activity and
conditioned place preference in rats. Psychopharmacol-
ogy (Berl) 2004;174:389-95.
96. DiFranza JR, Rigotti NA, McNeill AD, Ockene JK,
Savageau JA, St Cyr D, et al. Initial symptoms of
nicotine dependence in adolescents. Tob Control 2000;
9:313-9.
97. Vastola BJ, Douglas LA, Varlinskaya EI, Spear LP.
Nicotine-induced conditioned place preference in ado-
lescent and adult rats. Physiol Behav 2002;77:107-14.
98. Breslau N, Johnson EO. Predicting smoking cessation
and major depression in nicotine-dependent smokers.
Am J Public Health 2000;90:1122-7.
99. Hymowitz N, Cummings KM, Hyland A, Lynn WR,
Pechacek TF, Hartwell TD. Predictors of smoking ces-
sation in a cohort of adult smokers followed for ve
years. Tob Control 1997;6(Suppl 2):S57-62.
100. Ezzati M, Lopez AD. Estimates of global mortality
attributable to smoking in 2000. Lancet 2003;362:847-
52.
101. McBride PE. The health consequences of smoking. Car-
diovascular diseases. Med Clin North Am 1992;76:333-
53.
102. Peto R, Darby S, Deo H, Silcocks P, Whitley E, Doll R.
Smoking, smoking cessation, and lung cancer in the UK
since 1950: combination of national statistics with two
case-control studies. BMJ 2000;321:323-9.
103. Sellers EM, Tyndale RF, Fernandes LC. Decreasing
smoking behaviour and risk through CYP2A6 inhibi-
tion. Drug Discov Today 2003;8:487-93.
104. Silagy C, Mant D, Fowler G, Lancaster T. Nicotine
replacement therapy for smoking cessation. Cochrane
Database Syst Rev 2000;CD000146.
105. Fiore MC, Smith SS, Jorenby DE, Baker TB. The ef-
fectiveness of the nicotine patch for smoking cessation.
A meta-analysis. JAMA 1994;271:1940-7.
106. Patten CJ, Peterson LA, Murphy SE. Evidence for met-
abolic activation of N-nitrosonornicotine and
N-nitrosobenzylmethylamine by a rat nasal coumarin
hydroxylase. Drug Metab Dispos 1998;26:177-80.
107. Patten CJ, Smith TJ, Friesen MJ, Tynes RE, Yang CS,
Murphy SE. Evidence for cytochrome P450 2A6 and
3A4 as major catalysts for N-nitrosonornicotine alpha-
hydroxylation by human liver microsomes. Carcinogen-
esis 1997;18:1623-30.
108. Smith GB, Bend JR, Bedard LL, Reid KR, Petsikas D,
Massey TE. Biotransformation of 4-(methylnitros-
amino)-1-(3-pyridyl)-1-butanone (NNK) in peripheral
CLINICAL PHARMACOLOGY & THERAPEUTICS
156 Malaiyandi, Sellers, and Tyndale MARCH 2005
human lung microsomes. Drug Metab Dispos 2003;
31:1134-41.
109. Yamazaki H, Inui Y, Yun CH, Guengerich FP, Shimada
T. Cytochrome P450 2E1 and 2A6 enzymes as major
catalysts for metabolic activation of N-nitroso-
dialkylamines and tobacco-related nitrosamines in hu-
man liver microsomes. Carcinogenesis 1992;13:1789-
94.
110. Law MR, Morris JK, Watt HC, Wald NJ. The dose-
response relationship between cigarette consumption,
biochemical markers and risk of lung cancer. Br J
Cancer 1997;75:1690-3.
111. Kamataki T, Nunoya K, Sakai Y, Kushida H, Fujita K.
Genetic polymorphism of CYP2A6 in relation to can-
cer. Mutat Res 1999;428:125-30.
112. Miyamoto M, Umetsu Y, Dosaka-Akita H, Sawamura
Y, Yokota J, Kunitoh H, et al. CYP2A6 gene deletion
reduces susceptibility to lung cancer. Biochem Biophys
Res Commun 1999;261:658-60.
113. Wang H, Tan W, Hao B, Miao X, Zhou G, He F, et al.
Substantial reduction in risk of lung adenocarcinoma
associated with genetic polymorphism in CYP2A13, the
most active cytochrome P450 for the metabolic activa-
tion of tobacco-specic carcinogen NNK. Cancer Res
2003;63:8057-61.
114. Tan W, Chen GF, Xing DY, Song CY, Kadlubar FF,
Lin DX. Frequency of CYP2A6 gene deletion and its
relation to risk of lung and esophageal cancer in the
Chinese population. Int J Cancer 2001;95:96-101.
115. Sellers EM, Ramamoorthy Y, Zeman MV, Djordjevic
MV, Tyndale RF. The effect of methoxsalen on nicotine
and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone
(NNK) metabolism in vivo. Nicotine Tob Res 2003;5:
891-9.
116. Kharasch ED, Hankins DC, Taraday JK. Single-dose
methoxsalen effects on human cytochrome P-450 2A6
activity. Drug Metab Dispos 2000;28:28-33.
117. Koenigs LL, Peter RM, Thompson SJ, Rettie AE,
Trager WF. Mechanism-based inactivation of human
liver cytochrome P450 2A6 by 8-methoxypsoralen.
Drug Metab Dispos 1997;25:1407-15.
118. Sellers EM, Tyndale RF. Mimicking gene defects to
treat drug dependence. Ann N Y Acad Sci 2000;909:
233-46.
119. Takeuchi H, Saoo K, Yokohira M, Ikeda M, Maeta H,
Miyazaki M, et al. Pretreatment with 8-methoxy-
psoralen, a potent human CYP2A6 inhibitor, strongly
inhibits lung tumorigenesis induced by 4-(methylnitros-
amino)-1-(3-pyridyl)-1-butanone in female A/J mice.
Cancer Res 2003;63:7581-3.
120. Sellers EM, Kaplan HL, Tyndale RF. Inhibition of
cytochrome P450 2A6 increases nicotines oral bio-
availability and decreases smoking. Clin Pharmacol
Ther 2000;68:35-43.
121. Ariyoshi N, Sawamura Y, Kamataki T. A novel single
nucleotide polymorphism altering stability and activity
of CYP2a6. Biochem Biophys Res Commun 2001;281:
810-4.
122. Xu C, Rao YS, Xu B, Hoffmann E, Jones J, Sellers EM,
et al. An in vivo pilot study characterizing the new
CYP2A6*7, *8, and *10 alleles. Biochem Biophys Res
Commun 2002;290:318-24.
122a.Mwenifumbo JC, Myers MG, Wall T, Lin SK, Sellers
EM, Tyndale RF. Ethnic variation in CYP2A6*7,
CYP2A6*8 and CYP2A6*10 as assessed with a novel
haplotyping method. Pharmacogenetics. In press 2005.
123. Hadidi H, Zahlsen K, Idle JR, Cholerton S. A single
amino acid substitution (Leu160His) in cytochrome
P450 CYP2A6 causes switching from 7-hydroxylation
to 3-hydroxylation of coumarin. Food Chem Toxicol
1997;35:903-7.
124. Oscarson M, Gullsten H, Rautio A, Bernal ML, Sinues
B, Dahl ML, et al. Genotyping of human cytochrome
P450 2A6 (CYP2A6), a nicotine C-oxidase. FEBS Lett
1998;438:201-5.
125. Ariyoshi N, Takahashi Y, Miyamoto M, Umetsu Y,
Daigo S, Tateishi T, et al. Structural characterization of
a new variant of the CYP2A6 gene (CYP2A6*1B)
apparently diagnosed as heterozygotes of CYP2A6*1A
and CYP2A6*4C. Pharmacogenetics 2000;10:687-93.
126. Nunoya KI, Yokoi T, Kimura K, Kainuma T, Satoh K,
Kinoshita M, et al. A new CYP2A6 gene deletion re-
sponsible for the in vivo polymorphic metabolism of
()-cis-3,5-dimethyl-2-(3-pyridyl)thiazolidin-4-one hy-
drochloride in humans. J Pharmacol Exp Ther 1999;
289:437-42.
127. Oscarson M, McLellan RA, Gullsten H, Yue QY, Lang
MA, Bernal ML, et al. Characterisation and PCR-based
detection of a CYP2A6 gene deletion found at a high
frequency in a Chinese population. FEBS Lett 1999;
448:105-10.
128. Nunoya K, Yokoi T, Kimura K, Inoue K, Kodama T,
Funayama M, et al. A new deleted allele in the human
cytochrome P450 2A6 (CYP2A6) gene found in indi-
viduals showing poor metabolic capacity to coumarin
and ()-cis-3,5-dimethyl-2-(3-pyridyl)thiazolidin-4-
one hydrochloride (SM-12502). Pharmacogenetics
1998;8:239-49.
129. Oscarson M, McLellan RA, Gullsten H, Agundez JA,
Benitez J, Rautio A, et al. Identication and characteri-
sation of novel polymorphisms in the CYP2A locus:
implications for nicotine metabolism. FEBS Lett 1999;
460:321-7.
130. Kitagawa K, Kunugita N, Kitagawa M, Kawamoto T.
CYP2A6*6, a novel polymorphism in cytochrome p450
2A6, has a single amino acid substitution (R128Q) that
inactivates enzymatic activity. J Biol Chem 2001;276:
17830-5.
131. Yoshida R, Nakajima M, Watanabe Y, Kwon JT, Yokoi
T. Genetic polymorphisms in human CYP2A6 gene
causing impaired nicotine metabolism. Br J Clin Phar-
macol 2002;54:511-7.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):145-58 CYP2A6 genetic variation and smoking 157
132. Daigo S, Takahashi Y, Fujieda M, Ariyoshi N,
Yamazaki H, Koizumi W, et al. A novel mutant allele of
the CYP2A6 gene (CYP2A6*11) found in a cancer
patient who showed poor metabolic phenotype towards
tegafur. Pharmacogenetics 2002;12:299-306.
133. Oscarson M, McLellan RA, Asp V, Ledesma M, Ruiz
ML, Sinues B, et al. Characterization of a novel
CYP2A7/CYP2A6 hybrid allele (CYP2A6*12) that
causes reduced CYP2A6 activity. Hum Mutat 2002;20:
275-83.
134. Kiyotani K, Yamazaki H, Fujieda M, Iwano S, Mat-
sumura K, Satarug S, et al. Decreased coumarin
7-hydroxylase activities and CYP2A6 expression levels
in humans caused by genetic polymorphism in CYP2A6
promoter region (CYP2A6*9). Pharmacogenetics 2003;
13:689-95.
135. Pitarque M, von Richter O, Oke B, Berkkan H, Oscar-
son M, Ingelman-Sundberg M. Identication of a single
nucleotide polymorphism in the TATA box of the
CYP2A6 gene: impairment of its promoter activity.
Biochem Biophys Res Commun 2001;284:455-60.
136. Yoshida R, Nakajima M, Nishimura K, Tokudome S,
Kwon JT, Yokoi T. Effects of polymorphism in pro-
moter region of human CYP2A6 gene (CYP2A6*9) on
expression level of messenger ribonucleic acid and en-
zymatic activity in vivo and in vitro. Clin Pharmacol
Ther 2003;74:69-76.
137. Paschke T, Rieer M, Schuler-Metz A, Wolz L, Scherer
G, McBride CM, et al. Comparison of cytochrome P450
2A6 polymorphism frequencies in Caucasians and
African-Americans using a new one-step PCR-RFLP
genotyping method. Toxicology 2001;168:259-68.
138. Chen GF, Tang YM, Green B, Lin DX, Guengerich FP,
Daly AK, et al. Low frequency of CYP2A6 gene poly-
morphism as revealed by a one-step polymerase chain
reaction method. Pharmacogenetics 1999;9:327-32.
CLINICAL PHARMACOLOGY & THERAPEUTICS
158 Malaiyandi, Sellers, and Tyndale MARCH 2005
PHARMACOKINETICS AND
DRUG DISPOSITION
Influence of menstrual cycle on cytochrome
P450 2A6 activity and cardiovascular effects
of nicotine
Background and Objectives: The phase of the menstrual cycle has been reported to affect frequency of smok-
ing, withdrawal symptoms, and the likelihood of smoking cessation in women. Cytochrome P450 (CYP) 2A6
is primarily responsible for the metabolism of nicotine. Our objective was to evaluate the effect of the phase
of the menstrual cycle on the activity of CYP2A6 and the cardiovascular effects of nicotine.
Methods: Eleven healthy, nonsmoking women received a 30-minute combined infusion of deuterium-labeled
nicotine and cotinine (0.5 g kg
1
min
1
of each compound) during the midfollicular and midluteal
phases of the menstrual cycle. Nicotine and cotinine pharmacokinetic parameters and plasma adrenocortico-
tropic hormone (ACTH), epinephrine, and norepinephrine responses were measured over time.
Results: There were no biologically or statistically significant differences in the comparison of menstrual cycle
phases with regard to the pharmacokinetics of nicotine and cotinine. Nicotine clearance was 1000 315
mL/min and 1047 271 mL/min in the follicular and luteal phases, respectively (geometric mean ratio,
1.06; 90% confidence interval, 0.87-1.29). Cotinine clearance was 44 20 mL/min and 55 42 mL/min
in the follicular and luteal phases, respectively (geometric mean ratio, 1.13; 90% confidence interval, 0.90-
1.41). Nicotine infusion increased blood pressure, heart rate, and epinephrine concentrations. There were no
differences in catecholamine, ACTH, or hemodynamic responses to nicotine infusion between menstrual cycle
phases, although norepinephrine concentrations were constantly higher in the luteal phase compared with the
follicular phase.
Conclusions: CYP2A6 activity is not affected by menstrual cycle phase, and it is unlikely that menstrual
cyclerelated smoking habits of women are determined by changes in nicotine pharmacokinetics. The effects
of nicotine on plasma ACTH and catecholamine levels and hemodynamic parameters are not altered by
menstrual cycle phase in healthy, nonsmoking women. (Clin Pharmacol Ther 2005;77:159-69.)
Janne Hukkanen, MD, PhD, Steven G. Gourlay, MBBS, PhD,
Saraswati Kenkare, PhD, and Neal L. Benowitz, MD San Francisco and South San Francisco,
Calif
From the Division of Clinical Pharmacology and Experimental Thera-
peutics, Medical Service, San Francisco General Hospital Medical
Center, and Departments of Medicine, Psychiatry, and Biopharma-
ceutical Sciences, University of California, San Francisco, San Fran-
cisco, and Pharmacological Sciences, Genentech Inc, South San Fran-
cisco.
This study was presented in part at the annual meeting of the
American Society for Clinical Pharmacology and Therapeutics
(abstract MPI-107), Atlanta, Ga, March 24-27, 2002.
Supported by US Public Health Service grants DA02277 and
DA12393 from the National Institute on Drug Abuse, National
Institutes of Health, and grants from the Academy of Finland, the
Paavo Nurmi Foundation, and the Finnish Cultural Foundation to
J.H. Carried out in part at the General Clinical Research Center at
San Francisco General Hospital Medical Center with the support of
the Division of Research Resources, National Institutes of Health
(5-MO1-RR-00083-41).
Received for publication July 30, 2004; accepted Oct 24, 2004.
Reprint requests: Neal L. Benowitz, MD, Division of Clinical Phar-
macology and Experimental Therapeutics University of California,
San Francisco, Box 1220, San Francisco, CA 94143-1220.
E-mail: nbeno@itsa.ucsf.edu
0009-9236/$30.00
Copyright 2005 by the American Society for Clinical Pharmacology
and Therapeutics.
doi:10.1016/j.clpt.2004.10.012
159
The phase of the menstrual cycle inuences the smok-
ing habits of women; the frequency of smoking is slightly
increased during the late luteal phase.
1-3
Nicotine with-
drawal symptoms are also greater in the late luteal
phase,
4-8
and smoking cessation is reported to be harder
for female smokers in premenstruum compared with
women in midcycle and men.
9,10
The transdermal nicotine
patch has a greater effect in reducing craving during
smoking abstinence in the late luteal phase compared with
the follicular phase.
7
These observations suggest that hor-
monal changes during the menstrual cycle affect the
smoking behavior of women.
Changes in the metabolism of nicotine have been
shown to affect smoking behavior. Lower cytochrome
P450 (CYP) 2A6 activity caused by the presence of
defective alleles of CYP2A6, as well as pharmacologic
inhibition of CYP2A6, is associated with reduced cig-
arette consumption.
11,12
The main pathway of nicotine
metabolism is C-oxidation to cotinine, which accounts
for 70% to 80% of nicotine metabolism on average.
13,14
Cotinine is further metabolized to trans-3-hydroxy-
cotinine. Both of these reactions are catalyzed by he-
patic CYP2A6.
15,16
The inuence of sex hormones on nicotine metabo-
lism is suggested by the ndings that female gender,
pregnancy, and the use of oral contraceptives accelerate
the metabolism of nicotine and cotinine. Clearance is
increased by 60% and 140% for nicotine and cotinine,
respectively, in pregnancy compared with post partum
as a result of a rise in metabolic clearance.
17
A study
comparing women during pregnancy and again post
partum found that mean salivary cotinine concentration
per cigarette was higher when not pregnant (3.5 ng/mL
versus 9.9 ng/mL), consistent with higher cotinine
clearance during pregnancy.
18
Pregnant smokers had a
substantially lower level of serum nicotine than ex-
pected when standardized for their nicotine intake com-
pared with population-based values.
19
These results
suggest that CYP2A6 is induced during pregnancy,
possibly by sex hormones. Results from a recently
completed twin study with intravenous infusions of
both nicotine and cotinine show that nicotine and coti-
nine clearances are higher in women compared with
men and oral contraceptive use further accelerates nic-
otine and cotinine clearances in women.
20
Nicotine
clearance and cotinine clearance were 13% and 26%
higher, respectively, in women not taking oral contra-
ceptives compared with men. Oral contraceptive use
induced nicotine and cotinine clearances by 30% and
33%, respectively, compared with women not taking
oral contraceptives. The gender difference was also
detected in a recent study on smokers showing that the
ratio of nicotine to cotinine plus 3-hydroxycotinine in
24-hour urine collection is signicantly lower in
women, indicating faster metabolism in women com-
pared with men.
21
Thus there are reasons to suspect that the high estro-
gen and progesterone concentrations in the luteal phase
of the menstrual cycle might accelerate the metabolism
of nicotine. The effect of the menstrual cycle on
CYP2A6 activity has not been studied previously. The
aims of our study were to determine the effects of the
menstrual cycle on the disposition kinetics of nicotine
and to test the hypothesis that CYP2A6 activity is
induced in the luteal phase.
We also measured the effects of nicotine on cardio-
vascular parameters and hormonal responses in the
follicular and luteal phases of the menstrual cycle.
Nicotine is a sympathomimetic drug that causes cate-
cholamine release, increases heart rate and blood pres-
sure, and constricts some blood vessels. These hemo-
dynamic effects of nicotine may contribute to tobacco-
induced cardiovascular disease.
22
Nicotine-mediated
release of adrenocorticotropic hormone (ACTH) has
been speculated to contribute to the development of
addiction.
23
The menstrual cycle can also affect nor-
epinephrine concentrations and some hemodynamic re-
sponses to sympathomimetics.
24-26
No previous study
has examined the combined effect of menstrual cycle
phase and nicotine infusion on these parameters.
METHODS
Subjects. Healthy, premenopausal, nonsmoking
women were recruited for the study. The subjects used
no medications; they had not taken oral contraception
for the previous 3 months and had taken no other
prescription medication for the previous 28 days. Over-
the-counter medications had not been taken for the
previous 3 days. Exclusion criteria included an irregu-
lar menstrual cycle (dened as menstrual cycle duration
of 25 days or35 days or intermittent bleeding),
pregnancy, drug or alcohol abuse, and abnormal liver or
renal function (dened as elevated serum creatinine
level or abnormal liver function tests). Plasma nicotine
and cotinine concentrations were measured at screening
to conrm nonsmoking. The study was approved by the
Committee on Human Research at the University of
California, San Francisco (San Francisco, Calif). Writ-
ten, informed consent was obtained from each subject.
The subjects were nancially compensated for
participation.
The number of subjects (N11) was based on a power
analysis for paired Student t test of nicotine clearance to
CLINICAL PHARMACOLOGY & THERAPEUTICS
160 Hukkanen et al MARCH 2005
detect an effect size of 25% assuming a coefcient of
variation of 25% with .05 and .2.
Experimental protocol. Subjects were admitted
twice to the General Clinical Research Center at San
Francisco General Hospital Medical Center (San Fran-
cisco, Calif)once during the follicular phase and
once during the luteal phase. Serum estradiol and pro-
gesterone concentrations were measured during each
admission to conrm the phase of menstrual cycle. Day
1 of the follicular phase was determined by the onset of
menstrual bleeding. The day of ovulation was deter-
mined by a positive luteinizing hormone surge detected
by use of a home diagnostic ovulation kit (OvuKit;
Quidel Corporation, San Diego, Calif) and was speci-
ed as day 1 of the luteal phase. The day of admission
was assigned by counting the days from the onset of
menstruation (day 5-7 for the follicular phase) and by
counting the days after ovulation (day 6-8 for the luteal
phase). Subjects were instructed to test the rst morn-
ing urine with the ovulation kit 4 days before the
predicted midcycle and to continue the use of the kit
every morning (at least 9 days) until a positive result
was detected. The sequence (ie, follicular or luteal
phase rst) of the 2 admissions was alternated to con-
trol for any inuence of the order of testing. Subjects
were disqualied if hormonal measurements or ovula-
tion kit results were inconsistent with a normal cycle.
(If the rst admission was planned to be in the luteal
phase and the subject did not ovulate, she was disqual-
ied. If the rst admission was in the follicular phase
and the subject did not ovulate 2 consecutive times, she
was disqualied.)
The study was conducted as an outpatient study. Sub-
jects were asked to come to the clinical research center at
7 AM after fasting overnight. Two intravenous catheters
were inserted, one in the nondominant forearm for blood
drawing and another in the dominant antecubital vein for
infusion, and samples for estradiol and progesterone mea-
surements were taken. Subjects received an intravenous
infusion of deuterium-labeled nicotine and cotinine (0.5
g kg
1
min
1
of nicotine-d
2
[3,3-dideuteronicotine]
and 0.5 g kg
1
min
1
of cotinine-d
4
[2,4,5,6-tetra-
deuterocotinine]) for 30 minutes. The dose for 1 subject
was calculated erroneously, resulting in a lower infusion
rate of 0.4 g kg
1
min
1
. The same dose was infused
during both menstrual phases for all patients. Blood pres-
sure, heart rate, and nger skin temperature were mea-
sured at 15, 10, 5, 0, 5, 10, 15, 20, 25, 30, 40, 45, 50,
60, and 90 minutes and then at 2, 3, and 4 hours after the
beginning of the infusion. Blood samples for nicotine and
cotinine measurements were drawn at 0, 15, 30, 45, 60,
and 90 minutes and then at 2, 3, 4, 6, 8, 24, 48, 72, and 96
hours after the start of the infusion. Blood samples for
measurements of norepinephrine and epinephrine concen-
trations were taken at 30, 15, 0, 15, and 30 minutes,
and blood samples for ACTHmeasurements were taken at
30, 15, 0, 15, 30, 45, 60, and 90 minutes. Urine was
collected for 8 hours after infusion. Subjects were permit-
ted to sit up or to stand, and a meal was served 1 hour after
the infusion was completed. Subjects were discharged 8
hours after the start of the infusion. Subjects came back to
the study center on the next 4 days to have blood samples
taken.
Analytic methods. Nicotine and metabolite concentra-
tions in plasma were determined by gas chromatography
mass spectrometry.
27
Nicotine and metabolite concentra-
tions in urine were determined by liquid chromatography
tandem mass spectrometry.
28
Glucuronide-conjugated
nicotine, cotinine, and trans-3-hydroxycotinine levels
were measured as the difference in the total concentration
before and after hydrolysis by incubation with 2000 U of
-glucuronidase from Escherichia coli (Sigma, St Louis,
Mo) per 1 mL of urine.
Plasma concentrations of epinephrine and norepi-
nephrine were determined by means of HPLC with a
reversed-phase column with electrochemical detection
as described previously.
29
Plasma ACTH measure-
ments were performed with a chemiluminescence im-
munometric assay with ACTH 100T Kit (Nichols In-
stitute Diagnostics, San Juan Capistrano, Calif). Serum
estradiol and progesterone measurements were per-
formed with automated immunoassay by Unilab Cor-
poration (Tarzana, Calif).
Data analysis. Pharmacokinetic parameters were es-
timated from the plasma concentrations by use of
model-independent methods described previously
13
with the WinNonlin pharmacokinetic analyses package
(Pharsight Corp, Mountain View, Calif). Total clear-
ance was :
CL
nic
Dose nicd
2
AUC nicd
2
CL
cot
Dose cotd
4
AUC cotd
4
where CL is clearance (in milliliters per minute), AUC
is the area under the plasma concentrationtime curve
extrapolated to innity, nic is nicotine, and cot is coti-
nine. Renal clearance (CL
renal
) was calculated as fol-
lows:
CL
renal
, half-life; V
ss
, steady-state volume of distribution.
CLINICAL PHARMACOLOGY & THERAPEUTICS
162 Hukkanen et al MARCH 2005
3-fold in the luteal phase compared with the follicular
phase. Exclusion of this subject did not change the
results. The sequence of the 2 treatments (follicular or
luteal phase rst) did not inuence the results. Mean
intraindividual coefcients of variation for clearance
were 15.3% and 17.5% for nicotine-d
2
and cotinine-d
4
,
respectively. Mean serum estradiol concentrations were
89.9 63.5 pg/mL and 194.0 52.1 pg/mL for the
follicular and luteal phases, respectively. Mean serum
progesterone concentrations were 0.4 0.2 ng/mL and
9.6 4.3 ng/mL, respectively. Estradiol and proges-
terone concentrations did not correlate with nicotine or
cotinine clearances.
Urine metabolite patterns. The values for percent-
age recovery of nicotine and its metabolites derived
from nicotine-d
2
are shown in Table III. No signicant
differences between phases were detected.
Table II. Effect of menstrual cycle phases on disposition kinetics of cotinine-d
4
(N 11)
Follicular
(mean SD)
Luteal
(mean SD)
Geometric mean ratios
(luteal/follicular) and
90% condence interval
C
max
(ng/mL) 25.2 8.4 22.3 7.3 0.89 (0.77-1.04)
t
, elimination half-life.
*P .01 for 300 mL versus 1200 mL for water or grapefruit juice.
P .001 for water versus grapefruit juice for 300 mL or 1200 mL at equivalent volume.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):170-7 Grapefruit juice, fexofenadine, and OATP 173
broad pH range. Consequently, it seems unlikely that
the acidic pH of grapefruit juice would decrease the
dissolution of fexofenadine in the gastrointestinal tract.
Moreover, a greater volume of juice might be expected
to favor dissolution of fexofenadine. Given that 1200
mL grapefruit juice diminished the oral bioavailability
of fexofenadine more than 300 mL grapefruit juice, it
seemed unlikely that this juice acted primarily by ad-
versely affecting dissolution of this drug.
Fexofenadine is eliminated essentially unchanged,
and the t
max
values of this drug were similar among
treatments.
6
Thus it is doubtful that chemical instabil-
ity, changed drug metabolism, and altered transit to the
intestinal site of absorption were causes for the inter-
action.
The physicochemical properties of fexofenadine of a
high degree of ionization and large polar surface area
would predict negligible or low passive intestinal per-
meability.
8,9
Because this process is traditionally con-
sidered to be of primary importance for the passage of
drug from the small intestine into the portal circulation,
fexofenadine might be envisaged to have minimal or
minor intestinal absorption. Moreover, fexofenadine
encounters efux transport by P-glycoprotein in the
enterocytes.
5
Yet the absolute oral bioavailability of
fexofenadine is estimated to be 33% in humans.
10
Re-
cent ndings indicate that there is substantial expres-
sion of OATP-A, which enables uptake transport of
fexofenadine, on the luminal membrane of human in-
testinal enterocytes.
11
In addition, OATP-B (OATP2B1)
has been noted to be capable of mediating the cellular
uptake of fexofenadine and is expressed in the intes-
tine.
12,13
However, we have not been able to show that
cells overexpressing OATP-B cause signicant uptake
transport of fexofenadine (R.B.K., unpublished data,
September 2004). Although it is possible that both
OATP-A and OATP-B may play a role in fexofenadine
absorption from the intestine, OATP-A appears to have
far greater afnity for fexofenadine and is most likely
the major determinant of the extent of absorption of
fexofenadine into the portal circulation.
In the liver, OATPs are expressed on the sinusoidal
membrane of hepatocytes.
3
However, the OATP-A iso-
form has not been detected there. P-glycoprotein is
located on the bile canalicular membrane. Given that
biliary secretion is a major route for elimination of
fexofenadine, this might reect the cooperative action
of another OATP isoform plus P-glycoprotein. How-
ever, verapamil and ketoconazole, potent inhibitors of
P-glycoprotein, noticeably enhanced the systemic
availability of fexofenadine apparently by selectively
impairing biliary excretion.
14,15
Thus the biliary excre-
tion of fexofenadine might be mainly dependent on
efux transport mediated by P-glycoprotein.
In this investigation, grapefruit juice decreased the
AUC and C
max
but did not affect the t
max
or elimination
t
F D[i,j 1] k
a
V[i](k
a
k
e
)
(exp[k
e
(t
j
t
j1
)]
exp[k
a
(t
j
t
j1
)]) (3)
We used the above-described iterative algorithm to
estimate the hourly plasma concentration levels based
on the dosing events recorded by the MEMS records.
Specically, we calculated the values of Cp[i,j] in
hourly increments after the rst dosing event for the
entire duration of the study period covered by the
MEMS logs. We then averaged the estimated hourly
Cp[i,j] in each study subject for the mean hourly
plasma concentration for the study period. We denoted
this average as Cp
ave
[i] for the ith subject as follows:
Cp
ave
J
i j1
J
i
Cp
i,j
, where J
i
was the num-
ber of hours contained in the subjects MEMS log.
Therefore Cp
ave
[i] can be viewed as an approximation
of the mean plasma concentration of metoprolol during
the study period.
In a similar fashion, we calculated the intended
plasma concentration level under the assumption of
perfect adherence to the prescribed dose and frequency
of administration. We denoted the average of hourly Cp
values under the assumption of perfect adherence as
Cp
ave
[i]. It should be noted that Cp
ave
[i] can also be
obtained by use of a steady-state equation.
19
The dif-
ference between Cp
ave
[i] and Cp
ave
[i], therefore, re-
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):189-201 Adherence measures for electronic dosing data 191
ected the deviation of the projected Cp from the
intended (or prescribed) level. The greater this devia-
tion, the poorer the patients adherence to the pre-
scribed metoprolol regimen. Hence we propose to use
Cp
ave
[i] Cp
ave
[i] Cp
ave
[i] to measure the med-
ication adherence of the ith patient. Unlike most of the
existing adherence measures, Cp
ave
[i] also accommo-
dates the pharmacokinetic properties of the drug. For a
drug with a longer half-life, the plasma concentration
has a slower decrement and Cp
ave
[i] is less sensitive
to a delayed administration. Fig 1 depicts the estimated
Cp curves of metoprolol for an abbreviated time period
in 4 representative participants.
An alternative adherence measure was the ratio
Cp
ave
[i]/Cp
ave
[i], which quantied the proportion of
intended plasma concentration achieved by the ith pa-
tient. A larger value for this ratio indicated better pa-
tient adherence to the prescribed regimen.
Metoprolol. At therapeutic doses, metoprolol acts by
selectively antagonizing the
1
receptor in the myocar-
dium, producing negative chronotropism and inotro-
pism. Long-term use of metoprolol in heart failure
patients has been shown to be benecial in reducing
hospitalization and mortality rates.
20
Abrupt discontin-
uation and erratic administration of the drug, on the
other hand, are known risk factors of rebound tachy-
cardia, hypertension, and subsequent cardiac isch-
emia.
21
Patients in this study took immediate-release
metoprolol at doses ranging from 12.5 mg to 100 mg at
a frequency of 12 hours.
We used a 1-compartment model with published
pharmacokinetic parameters to estimate plasma con-
Fig 1. Estimated and intended metoprolol plasma concentration curves for 4 study subjects during
abbreviated time interval for exemplication. The dashed curve depicts the time course of the
estimated plasma concentration under perfect adherence and thus represents the intended dosing
pattern. The continuous curve depicts the MEMS-recorded dosing pattern. The upper horizontal line
represents the overall average value of intended metoprolol exposure (Cp
ave
[i]), and the lower
horizontal line represents the hourly average metoprolol concentration (Cp
ave
[i]) estimated from the
MEMS-recorded dosing pattern.
CLINICAL PHARMACOLOGY & THERAPEUTICS
192 Tu et al MARCH 2005
centrations of metoprolol, although it more appropri-
ately follows a 2-compartment kinetic model. Two-
compartment models could be used when kinetic data
were available to t those models. Clearance based on
differences in metabolism varies between 50 L/h and
409 L/h. In the absence of individual data, we used the
population-based value of the elimination rate constant
(k
e
) and the 1-compartment model to estimate the
plasma concentration.
The pharmacokinetic parameters of metoprolol are
well established. The drug has an elimination half-life
of approximately 3.5 hours, bioavailability (F) of 0.5,
k
e
equal to 0.198/h, and time to peak Cp (t
max
) approx-
imately equal to 1.5.
22,23
On the basis of the following
equation, we solved for an estimate of the absorption
parameter (k
a
) equal to 1.585/h:
t
max
1
(k
a
k
e
)
ln
k
a
k
e
F d exp(k
a
) k
a
V[i](k
a
k
e
)
[exp(0.2k
e
) exp(0.2k
a
)]
F d k
a
V[i](k
a
k
e
)
[exp(1.2k
e
) exp(1.2k
a
)]
which again followed the standard 1-compartment
model.
At time t
3
2.0, we had d[i,3] 0 because the
patient did not take the medication at this time point;
the cumulative dose was as follows: D[i,3]
d [exp(2k
a
) exp(0.8k
a
)]. From equation 3, we
calculated the plasma concentration at this time point as
follows:
Cp[i,3]
F d k
a
V[i](k
a
k
e
)
[exp(1.2k
e
) exp(1.2k
a
)]
exp(0.8k
e
)
F d (exp(1.2k
a
) 1) k
a
V[i](k
a
k
e
)
[exp(0.8k
e
)
exp(0.8k
a
)]
F d k
a
V[i](k
a
k
e
)
[exp(2.0k
e
)
exp(2.0k
a
)]
F d k
a
V[i](k
a
k
e
)
[exp(0.8k
e
)
exp(0.8k
a
)]
where the rst term in the above-described expression
represented the portion of plasma concentration con-
tributed by the rst dosing event at time t
0
0 and the
second term represented the portion contributed by the
second dosing event at t
2
1.2. This showed how the
iterative algorithm was able to accommodate the cu-
mulative drug effect in multiple dose regimens.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):189-201 Adherence measures for electronic dosing data 201
BIBN4096BS antagonizes human
-calcitonin gene related peptideinduced
headache and extracerebral artery dilatation
Background and Objective: Calcitoningenerelatedpeptide (CGRP) plays a pivotal role inmigraine pathogenesis.
BIBN4096BS is the first CGRP receptor antagonist available for human studies, and its efficacy in the acute
treatment of migraine has been demonstrated. We investigated the ability of BIBN4096BS to inhibit human
CGRP (h-CGRP)induced headache and cerebral hemodynamic changes in healthy volunteers.
Methods: Ten healthy volunteers completed this double-blind, placebo-controlled crossover study with
2.5 mg BIBN4096BS and placebo as pretreatments before a 20-minute intravenous infusion of h-CGRP
(1.5 g/min). Transcranial Doppler ultrasonography was used to measure blood flow velocity in the middle
cerebral artery (MCA); regional and global cerebral blood flow(CBF) was measured by xenon 133 inhalation
single-photon emission computed tomography. The temporal and radial artery diameter was measured by
high-frequency ultrasound. Systemic hemodynamics, end-tidal partial pressure of carbon dioxide (PETCO
2
),
and headache were monitored.
Results: Of the 10 volunteers, 6 had a CGRP-induced headache during the in-hospital phase after placebo
pretreatment but none after BIBN4096BS (P .031). BIBN4096BS did not affect changes in the diameter
of the MCA or changes in CBF induced by h-CGRP. Vasodilatation of the extracranial arteries was,
however, significantly inhibited (P < .001 for temporal artery and P .001 for radial artery).
Conclusions: These results show that BIBN4096BS effectively prevents CGRP-induced headache and extra-
cerebral vasodilatation but does not significantly affect the induced cerebral hemodynamic changes. (Clin
Pharmacol Ther 2005;77:202-13.)
Kenneth A. Petersen, MD, Lisbeth H. Lassen, PhD, Steffen Birk, PhD,
Lynna Lesko, PhD, and Jes Olesen, DMSc Glostrup, Denmark, and Ridgefield, Conn
Calcitonin generelated peptide (CGRP) is a neu-
ropeptide found in the perivascular nerve terminals
surrounding arteries.
1
A measurable concentration of
CGRP is circulating in the blood at rest,
2
and CGRP
receptors are localized throughout the body.
3
Cerebral
and other cephalic arteries have a particularly rich
innervation of CGRP-containing afferent trigeminal
nerve bers, and these arteries, as studied in tissue
baths, are particularly sensitive to CGRP.
4
CGRP is
found in an increased concentration in external jugular
venous blood but not in blood from the cubital vein
during a migraine attack.
5
After infusion in patients
with migraine, CGRP caused a migraine-like headache
and in some a genuine migraine attack with associated
symptoms that were indistinguishable from the pa-
tients normal migraine attacks.
6
Peptide antagonists of
CGRP receptors have been available for experimental
studies for several years, but previously available com-
pounds have not been tested for safety and are, there-
fore, not suitable for human clinical studies. One CGRP
receptor antagonist, BIBN4096BS, has been developed
with the purpose of treating acute migraine, and a phase
II study has provided proof of efcacy.
7
Although
BIBN4096BS potently interacts with the human CGRP
From the Danish Headache Center, University of Copenhagen, and
Department of Neurology, Glostrup University Hospital, Glostrup,
and Boehringer Ingelheim Pharmaceuticals Inc, Ridgeeld.
Boehringer Ingelheim sponsored the study and provided
BIBN4096BS. The authors were independently responsible for the
study design, data analysis, and manuscript. The technical equip-
ment used was partly sponsored by the Villum Kann Rasmussen
Foundation, Toyota Foundation, and Simon Fougner Hartmann
Foundation. The Lundbeck Foundation funds the research of the
Danish Headache Center.
Received for publication June 29, 2004; accepted Oct 6, 2004.
Reprint requests: Kenneth A. Petersen, MD, Danish Headache Center,
University of Copenhagen and Department of Neurology, Glostrup
University Hospital, KAS Glostrup, DK-2600 Glostrup, Denmark.
E-mail: kapetersen@dadlnet.dk
0009-9236/$30.00
Copyright 2005 by the American Society for Clinical Pharmacology
and Therapeutics.
doi:10.1016/j.clpt.2004.10.001
202
receptor in vitro,
8,9
no information is available on its
ability to inhibit CGRP changes in human volunteers at
increased levels of the peptide as seen during a mi-
graine attack.
5
We, therefore, decided to conduct a placebo-
controlled, double-blind crossover experiment, in
which BIBN4096BS and placebo in random order were
used as pretreatment followed by infusion of human
CGRP (h-CGRP). The aims of this study were to
validate the ability of h-CGRP to provoke headache
and to describe its effect on cerebral, extracerebral, and
systemic circulatory parameters. Furthermore, we ana-
lyzed whether BIBN4096BS could partly or fully block
the changes evoked by h-CGRP. It was our hope that
the study could thus contribute to a better understand-
ing of the role of CGRP in neurovascular headache and
the site of action of this new antimigraine compound.
METHODS
Design and participants. This was a placebo-
controlled, double-blind crossover study that included
11 healthy subjects (7 men and 4 women). One female
participant had claustrophobia that developed during
baseline single-photon emission computed tomography
(SPECT) scanning and was excluded before any trial
medication was given. Ten participants completed both
treatment days. The participants were aged 24 to 31
years (mean, 26.5 years) and weighed 68 to 89.4 kg
(mean, 77.4 kg). The participants had no current or
previous cardiovascular, cerebrovascular, endocrine, or
neurologic disorder, including no migraine, hypoten-
sion, or hypertension. A frequency of tension-type
headache of 4 d/mo or lower was accepted. On the day
of enrollment, physical and neurologic examination,
electrocardiography, and blood sampling were done.
The healthy volunteers were randomized to receive
either 2.5 mg BIBN4096BS or placebo (xylitol 5%) as
an intravenously administered pretreatment of 10 min-
utes duration. After a free interval of 10 minutes,
1.5 g/min h-CGRP was administered continuously
for 20 minutes on both trial days. The 2 trial days were
separated by at least 1 week.
Boehringer Ingelheim GmbH supplied BIBN4096BS
and performed the randomization and blinding, which
was balanced (ClinPro, version 6; Clinical Systems,
Inc, Garden City, NY). The dose effective in the treat-
ment of acute migraine attacks was used.
7
Human-CGRP was purchased from Clinalfa AG,
Lufelngen, Switzerland. In a study performed previ-
ously, we used a dose of 2 g/min.
6
This dose, how-
ever, induced pronounced hypotension that in 2 patients
necessitated premature termination of the infusion. In
the current study we, therefore, used the lower dose of
1.5 g/min.
All participants gave written informed consent be-
fore randomization. The Ethical Committee of Copen-
hagen (KA00079gs) and the Danish Medicines Agency
(2612-1376) approved the study, which was conducted
in accordance with the Helsinki II Declaration and the
Guidelines for Good Clinical Practice.
10
Recording of adverse events. Every 15th minute
from time (T) zero (T
0
) (baseline) to T
240
(end of study
period), the volunteers were questioned regarding the
presence of adverse events (AEs) and rated headache.
Between questionings, the participants self-reported
any changes that they might have. The intensity of the
AEs was graded as mild, moderate, or severe, and their
relationship to study medication was classied as re-
lated or not related by the investigator. Headache in-
tensity was scored on an 11-point verbal rating scale
with 0 indicating no headache; 1 indicating a feeling of
the occurrence of something unusual inside the head,
not necessarily actual pain; 5 indicating headache of
medium severity; and 10 indicating worst imaginable
headache. Accompanying symptoms were recorded ac-
cording to the International Headache Classication.
11
During the study period, the investigator recorded the
AEs. After discharge, the volunteers made an hourly
recording of AEs up to 24 hours after the infusion of
placebo or BIBN4096BS.
Cerebral blood ow measurements. Global and re-
gional cerebral blood ow (CBF) was measured with
xenon 133 inhalation and SPECT with a brain-
dedicated camera (Ceraspect; DSI, Waltham, Mass).
The apparatus consisted of a stationary annular sodium
iodide crystal and a fast-rotating collimator system.
Each rotation took 10 seconds, thereby acquiring 1
frame in a 30-frame dynamic protocol of
133
Xe inha-
lation, with 3 background, 9 wash-in, and 18 wash-out
frames by use of the Kanno-Lassen algorithm.
12
A
photoelectric window of 70 to 100 keV was used.
Thirty-two slices were reconstructed in a 64 64
matrix with each pixel measuring 0.33 0.33 cm by
use of a Butterworth one-dimensional lter (cutoff, 1.5;
order, 6). The 32 slices were reduced to sets of 8
transaxial slices generated by adding 4 slices together
to a total slice thickness of 1.32 cm. A correction by use
of the Chang algorithm (m 0.05 cm) and nose
artifact was performed. The output for each pixel was
the inhibition constant (K
i
) value, and ow values were
estimated from these by use of the partition coefcient
() of 0.85 (gray matter).
A Datex Normocap 200 (Dameca, Roedovre, Den-
mark) was used for end-tidal partial pressure of carbon
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):202-13 CGRP receptor antagonism and vascular headache 203
dioxide (PETCO
2
) measurements during the CBF acqui-
sitions. A Ceratronic XAS SM 32C (Randers, Den-
mark) was used for the
133
Xe administration. Each
measurement lasted 5 minutes.
Calculations of ow in the perfusion territories of the
major cerebral arteries were performed by tting of
standard vascular regions of interest on the 5 rostral
slices at 3.6, 5.0, 6.3, 7.6, and 9 cm above the orbito-
meatal line. Flow in the territory of the middle cerebral
artery (MCA) (rCBF
MCA
) was calculated as a mean of
the left and right side.
Transcranial Doppler and C-scan. Transcranial
Doppler (TCD) ultrasonography (2 MHz) (Multi-Dop
X; DWL, Sipplingen, Germany) was used for the mea-
surement of blood ow velocity. The recordings were
done simultaneously and bilaterally as previously de-
scribed but with handheld probes.
13
Along the MCA, a
xed point was found for the measurement. The xed
point was chosen as close as possible to the bifurcation
of the anterior cerebral artery and MCA. The same x
point was used for each individual and for each record-
ing, for which the signal was optimized. On the basis of
the envelope curve (the spectral TCD curve), a time
averaged mean (V
mean
) over approximately 4 cardiac
cycles or 4 seconds was calculated by the built-in
software (version 7.40x of MDX TCD-7 software for
Multi-Dop X hardware, DWL). The nal measure used
for each time point was an average of 4 cycles (V
MCA
).
Simultaneously with the TCD recording, a mask cov-
ering the subjects mouth and nose region was placed
for the measurement of PETCO
2
(Datex Normocap 200;
Dameca).
A high-resolution ultrasound scanner, C-scan
(Dermascan C, 20 MHz; bandwidth, 15 MHz) (Cortex
Technology, Hadsund, Denmark),
14
was used to mea-
sure the diameter of the left temporal and left radial
artery. The diameter of the former was measured at the
front branch of the supercial temporal artery and the
latter at the wrist. To ensure that the repeated measure-
ments with TCD and C-scan were performed in the
same place, marks were drawn on the skin. After the
last recording on the rst trial day, the coordinates of
the marks were kept for reuse on the following trial
day.
Pharmacokinetics. Plasma concentrations of
BIBN4096BS were sampled at the following time
points: T
10
(baseline), T
9.5
, T
30
, T
60
, and T
180
on each
trial day in Vacutainer blood-collecting tubes with eth-
ylenediaminetetraacetic acid (K3 10-mL glasses;
Becton Dickinson, Rutherford, NJ). Samples were
stored on ice for a maximum of 30 minutes before
centrifuged for 10 minutes (2000 rpm) at 4C. The
plasma was stored at 20C until analyzed at Boehr-
inger Ingelheim Pharma GmbH & Co KG (Biberach an
der Riis, Germany). The plasma concentration of
CGRP was determined twice, at baseline (T
10
) and at
the end of the h-CGRP infusion (T
40
).
BIBN4096BS antibodies. BIBN4096BS was modi-
ed with succinic acid anhydride. This hapten was
covalently coupled to human serum albumin. Poly-
clonal antibodies were produced by immunization of
3-month-old female New Zealand rabbits with the im-
munogen in complete Freunds adjuvant. After several
booster immunizations, the antibodies were puried
from rabbit serum by use of protein ASepharose
(Sepharose is a registered trademark of Amersham
Biosciences).
BIBN4096BS analytic methods. The procedures
were conducted in accordance with current interna-
tional guidelines.
15
In this competitive enzyme-linked
immunosorbent assay, the biotinylated anti-
BIBN4096BS antibodies (immunoglobulin G fraction)
were bound to microtiter plates that were adsorptive-
coated with avidin. BIBN4096BS in the plasma sample
competed with added horseradish peroxidaselabeled
BIBN4096BS reagent for binding sites on the solid-
phase antibodies. After incubation, unbound
BIBN4096BS and plasma components were removed
by washing. Antibody-bound enzyme activity was de-
tected with a chromogenic substrate. The amount of
colored product formed was measured photometrically
and decreased with the increasing concentration of
BIBN4096BS in the plasma sample. The BIBN4096BS
concentration corresponding to the measured optical
absorbance was calculated via data tting of the non-
linear standard curve.
To compensate for slight variations in immuno-
chemical reaction parameters (such as temperature and
antibody binding capacity) between microplates, a stan-
dard curve was included on each plate. All steps of the
enzyme-linked immunosorbent assay were performed
at 22C 1C, which corresponded to room tempera-
ture of the air-conditioned laboratory.
Assay precision as assessed from 886 triplicate de-
terminations by construction of a precision prole was
9.1% coefcient of variance (CV) at the lower limit of
quantication, 2.7% CV at the upper limit of quanti-
cation, and 1.6% CV in the middle of the working range
(0.5 ng/mL).
Human-CGRP analysis. The analysis of
h-CGRP plasma concentrations was performed at the
Department of Clinical Physiology and Nuclear Medi-
cine, Glostrup Hospital (Glostrup, Denmark). The
method of analysis has been described in detail else-
CLINICAL PHARMACOLOGY & THERAPEUTICS
204 Petersen et al MARCH 2005
where.
2
In this study only 100 L of serum was used.
The normal values were 85 35.4 pmol/L for women
and 88 36.2 pmol/L for men (Schifter S, oral com-
munication, November 2002).
Trial procedure. The healthy volunteers began the
study at 8 AM, headache-free. For the preceding 8 hours,
they had abstained from drinking coffee, tea, and
caffeine-containing beverages and smoking tobacco
and they had not taken any medication, except oral
contraception. They rested in the supine position
throughout the study period (T
20
to T
180
) Two intra-
venous catheters (Optiva*2 [18 gauge]; Johnson &
Johnson, Ethicon SpA, Pomezia, Italy) were inserted
into the cubital veins, one for the administration of
human CGRP and BIBN4096BS and the other for
blood sampling. The volunteers rested for at least 30
minutes before baseline values of CBF, V
MCA
, tempo-
ral and radial diameter, blood pressure (BP), heart rate
(HR), and electrocardiogram were recorded. The start
of infusion of 2.5 mg BIBN4096BS or placebo was
designated as time zero (T
0
). The infusion lasted 10
minutes. At T
20
, a 20-minute infusion of h-CGRP (1.5
g/min) was initiated. Infusions were administered by
a time- and volume-controlled infusion pump (Braun
perfusor; B. Braun Melsungen AG, Melsungen,
Germany).
All measurements, except the CBF measurements,
were recorded quarterly for 3 hours (study period), and
BP, HR, electrocardiogram (Cardiofax; Nihon Kohden
Corporation, Tokyo, Japan), AEs, and headache were
recorded for an additional hour. BP and HR were
measured every 5 minutes for the rst hour and there-
after every 15th minute with an automatically inating
cuff (Omega 1400, In Vivo Research Laboratories Inc,
Copiague, NY). In the observation period from T
180
to
T
240
, the participants were allowed to sit upright. Three
SPECT scans were done as follows: at baseline, at T
60
,
and at T
90
. V
MCA
was measured immediately after each
SPECT scan.
The estimated perfusion (rCBF
x
) in the area of a
given artery (x) is dependent on the mean blood ow
velocity [V
mean(x)
] and the cross-sectional area (r
2
)
of the artery. The following equation is valid for the
regional CBF:
rCBF
(x)
Vmean
(x)
r
2
Hence,
Diameter
rCBF
2(x)
V
mean
2(x)
V
mean
1(x)
rCBF
1(x)
1
100
Diameter is the relative percentage change in diam-
eter, V
mean1(x)
is the mean blood velocity before infu-
sion of drugs, and V
mean2(x)
is the velocity at a relevant
time point after the infusion; the same designation is
applied for rCBF.
16,17
Statistics. Baseline was calculated as a mean of the
measurements at time points T
20
and T
10
in the
analysis. Values are presented as means SD. P .05
was considered signicant. All analyses were per-
formed by use of SPSS statistical software, version 10.0
(SPSS Inc, Chicago, Ill).
For changes over time on each trial day, V
MCA
,
global CBF, rCBF
MCA
, diameter of the temporal and
radial artery, BP, and PETCO
2
were analyzed by a uni-
variate ANOVA for the factors time and subject. If a
signicant change was found, a post hoc analysis
(Dunnett multiple comparisons test) was performed to
localize the change. To eliminate the risk of mass
signicance of measurements with numerous repeated
measurements, 4 points of interest were chosen as
follows: baseline, 45 minutes, 105 minutes, and 165
minutes. Absolute values were used for the statistical
analysis. For the comparison between BIBN4096BS
and placebo, a paired t test was performed for the
following measurements: V
MCA
, global CBF,
rCBF
MCA
, diameter of the temporal and radial arteries,
BP, and PETCO
2
. The summary measure for the t test
was the area under the curve (AUC) calculated on
percentage changes from baseline.
Immediate headache was dened as any headache
during the rst 60 minutes after the start of the
h-CGRP infusion. Any headache occurring thereafter
was referred to as delayed headache. Peak values and
area under the curve for headache (AUC
headache
) were
compared between the 2 trial days by use of the
Wilcoxon signed rank test. The occurrence of headache
and AEs on the 2 trial days was compared by use of the
McNemar test.
RESULTS
Baseline values. All baseline measurements of the
hemodynamic responses are summarized in Table I.
Only the baseline PETCO
2
measured simultaneously
with the TCD recordings showed a signicant differ-
ence between study days (P .03). The nding was
interpreted as incidental and was not taken into account
in the processing of data.
Effect of BIBN4096BS on CGRP-induced head-
ache and other AEs. On placebo days, 5 participants
had an immediate headache and 3 had a delayed head-
ache; the maximum immediate headache score was 2
and the maximum delayed headache score was 1. No
participants had an immediate headache but 1 had a
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):202-13 CGRP receptor antagonism and vascular headache 205
delayed headache after BIBN4096BS pretreatment.
The delayed headache occurred 6 hours after the infu-
sion of BIBN4096BS, lasted 3 hours, and was scored 1.
The effect of BIBN4096BS in preventing immediate
headache was signicant (P .034 for peak headache
and P .04 for AUC
headache
) and in preventing the
occurrence of any headache during the in-hospital
phase (P .031, McNemar test).
After placebo pretreatment, h-CGRP caused ush-
ing in all participants and all but 1 had bilateral con-
junctival injection. Eight experienced a sensation of
heat. Five reported palpitations. None of these CGRP-
induced changes were seen on days when
BIBN4096BS was administered as pretreatment (Table
II). AEs that could possibly be assigned to the CGRP
receptor antagonist were located to the infusion site.
CBF. Global CBF increased signicantly after
h-CGRP on both study days (P .007 after placebo
and P .009 after BIBN4096BS pretreatment). The
increase was measured 20 minutes after the h-CGRP
infusion was stopped. No difference was found be-
tween the 2 days (P .42).
After h-CGRP, rCBF
MCA
increased signicantly on
both trial days (P .003 and P .01), again 20
minutes after the h-CGRP infusion. No signicant
difference was observed between the 2 study days (P
.38). Data were not corrected for PETCO
2
, because no
signicant changes were found on either day (P .2
and P .6) or between treatment days (P .1).
TCD. V
MCA
did not vary signicantly over time
(P .3 for placebo and P .7 for BIBN4096BS), and
between the 2 trial days, no difference was seen (P
.74). On the basis of the rCBF
MCA
and V
MCA
measure-
ments, the effect on the relative percentage diameter
change of MCA can be estimated.
16,18
As seen in Table
III, a dilation of the MCA was found on both study
days. Compared with baseline, the dilation occurring on
placebo days reached signicance at T
60
(P .005).
This corresponded to a diameter increase of 9.3%
8.1%. When BIBN4096BS was given as pretreatment,
Table II. Effect of BIBN4096BS pretreatment on
h-CGRPinduced symptoms
Symptom
Placebo
plus
h-CGRP
(No.)
BIBN4096BS
(2.5 mg) plus
h-CGRP
(No.)
P value
(McNemar
test)
Flushing 10 0 P .002
Heat sensation 8 0 P .008
Palpitations 5 0 P .063
Conjunctival
injection
9 0 P .004
Headache 6 0 P .031
The ushing and conjunctival injection was based on the investigators
observations. Heat sensation and palpitation were reported and headache was
systematically scored. The data shown are from the entire in-hospital study
period.
Table I. Baseline values of measured variables
Measured variable
Placebo plus
h-CGRP
BIBN4096BS
(2.5 mg) plus
h-CGRP
P
value
Global CBF (mL 100 g brain tissue
1
min
1
) 46.7 10.8 45.6 10.5 .5
rCBF
MCA
(mL 100 g brain tissue
1
min
1
) 45.9 10.5 44.7 10.6 .3
PETCO
2
(mm Hg)
CBF 39 3.5 39 4.1 .9
TCD 41 3.0 39 3.6 .03*
V
MCA
(cm/s) 78 17.0 74 15.3 .1
C-scan
Temporal (mm) 1.26 0.4 1.29 0.3 .8
Radial (mm) 2.76 0.5 2.63 0.4 .2
Systolic blood pressure (mm Hg) 113 8 113 6 .9
Diastolic blood pressure (mm Hg) 64 8 64 5 .8
Mean arterial blood pressure (mm Hg) 79 6 80 7 .5
Heart rate (beats/min) 54 8 53 4 .7
Plasma CGRP (pmol/L) 89 20.3 91 20.4 .5
h-CGRP, Human -calcitonin generelated peptide; CBF, cerebral blood ow; rCBF
MCA
, cerebral blood ow in territory of middle cerebral artery; PETCO
2
, end-tidal
partial pressure of carbon dioxide; TCD, transcranial Doppler; V
MCA
, middle cerebral artery blood ow velocity (average of 4 cycles); CGRP, calcitonin generelated
peptide.
*Signicant difference between baseline on placebo and BIBN4096BS pretreatment days (paired t test).
CLINICAL PHARMACOLOGY & THERAPEUTICS
206 Petersen et al MARCH 2005
an increase was seen as well, but it did not reach
statistical signicance (P .2). There was no differ-
ence between placebo and BIBN4096BS pretreatment
(P .17). Data were not corrected for PETCO
2
, because
no signicant changes were found on the trial days
(P .4 on both days) or between treatment days (P
.45).
C-scan. A signicant change in temporal artery di-
ameter over time on placebo days was observed (P
.001). This signicance was seen at all time points after
the infusion of h-CGRP.
A signicant diameter change was not seen when
BIBN4096BS was infused as pretreatment (P .7).
The increase in temporal artery diameter was signi-
cantly inhibited by BIBN4096BS (P .001). The
radial artery revealed a similar nding, with a signi-
cant change on placebo days (P .01) and a nonsig-
nicant response after BIBN4096BS pretreatment
(P .07). The difference between placebo and
BIBN4096BS was signicant (P .001) (Fig 1).
Peripheral hemodynamics. Table IV summarizes
data on systolic BP, diastolic BP, mean arterial BP, and
HR. No signicant time-dependent changes were seen
in systolic, diastolic, or arterial mean BP on either trial
day or between days. The HR increased signicantly on
placebo days (P .001) at all time points compared
with baseline. This increase was not seen on
BIBN4096BS pretreatment days. The increase in HR
was signicantly inhibited by BIBN4096BS (P .003).
Pharmacokinetics of BIBN4096BSand h-CGRP. The
highest measured plasma concentration of
BIBN4096BS occurred just before the end of infusion
at 9.5 minutes after the start of the infusion, with a
mean of 170.4 25.4 ng/mL. The lowest plasma
concentration was measured 180 minutes after the start
of the infusion, with a mean of 8.7 5.7 ng/mL. No
BIBN4096BS was detected on placebo days (Fig 2).
The pharmacokinetic parameters for BIBN4096BS
have previously been published and are summarized.
On the basis of administration of 5 and 10 mg intrave-
nously for 10 minutes, BIBN4096BS had a total plasma
clearance of 12 L/h and a terminal half-life of 2.5 hours.
Approximately 15% of the dose was excreted un-
changed in urine, with a mean renal clearance of 2 L/h.
The geometric mean maximum plasma concentration
after 2.5 mg of BIBN4096BS was 210 ng/mL.
19
At the end of the CGRP administration, a signicant
difference between treatment days was found (P
.001, paired t test); the CGRP concentration was 342
68 pmol/L when placebo was administered as pretreat-
ment and 442 70 pmol/L when BIBN4096BS was
given (Fig 3). To elucidate whether this difference was
caused by a cross-reaction between BIBN4096BS and
our CGRP analytic kit, blood samples from 3 healthy
volunteers with the following concentrations of
BIBN4096BS were analyzed for CGRP: sample 1 and
2, 0 ng/mL (control); sample 3, 1000 ng/mL; sample 4,
500 ng/mL; sample 5, 250 ng/mL; and sample 6, 125
ng/mL. The analysis did not reveal any correlation
between the concentration of BIBN4096BS and CGRP.
Thus BIBN4096BS did not cross-react with CGRP
detection in our analytic kit.
DISCUSSION
This study has demonstrated that a specic CGRP
antagonist, BIBN4096BS, can prevent CGRP-induced
symptoms such as headache, ushing, heat sensation,
and palpitations. It furthermore prevents a CGRP-
induced increase in HR and dilatation of the supercial
temporal and radial artery, whereas it has no signicant
Table III. Effect of BIBN4096BS pretreatment on h-CGRPinduced cerebral hemodynamic changes
Time point
V
MCA
(cm/s)
Diameter
MCA
(%) and
Area
MCA
(%)
rCBF
MCA
(mL 100 g
1
min
1
)
Placebo
plus CGRP
BIBN4096BS
plus CGRP
Placebo
plus CGRP
BIBN4096BS
plus CGRP
Placebo
plus CGRP
BIBN4096BS
plus CGRP
Baseline 78 17.0 74 15.3 0 (0) 0 (0) 45.9 10.5 44.7 10.6
60 min 75 15.5 75 12.9 9.3 8.1*
(19.95 17.9)
3.8 5.4
(7.96 11.3)
52.4 9.4 49.1 12.1
90 min 77 14.3 74 15.5 3.4 7.5
(7.4 15.1)
3.0 7.9
(6.8 16.3)
49.3 12.7 47.9 10.1
Values are given as mean SD. The percentage change in the mean diameter (left and right sides) of the middle cerebral artery (MCA) was estimated according to
Dahl et al.
16
V
MCA
and rCBF on left and right sides analyzed separately showed similar results.
*Difference from baseline on 2 trial days (ANOVA, Dunnett post hoc): P .003.
Difference from baseline on 2 trial days (ANOVA, Dunnett post hoc): P .002.
Difference from baseline on 2 trial days (ANOVA, Dunnett post hoc): P .006.
Difference from baseline on 2 trial days (ANOVA, Dunnett post hoc): P .046.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):202-13 CGRP receptor antagonism and vascular headache 207
effect on CGRP-induced increase in CBF and dilatation
of the MCA.
Localization, function, and role of CGRP in mi-
graine. CGRP is one of the most potent vasodilators
known.
4
In the brain, immunohistochemical studies
have located the peptide to perivascular sensory
Table IV. Effect of BIBN4096BS pretreatment on h-CGRPinduced systemic hemodynamic changes
Time
point
Systolic blood pressure
(mm Hg)
Diastolic blood pressure
(mm Hg)
Mean blood pressure
(mm Hg)
Heart rate
(beats/min)
Placebo
plus CGRP
BIBN4096BS
plus CGRP
Placebo
plus CGRP
BIBN4096BS
plus CGRP
Placebo
plus CGRP
BIBN4096BS
plus CGRP
Placebo
plus CGRP
BIBN4096BS
plus CGRP
Baseline 113 8 113 6 64 8 64 5 79 6 80 7 54 8 53 4
45 min 115 9 111 9 59 5 67 7 80 6 81 7 72 8* 57 10
105 min 112 11 110 10 62 6 66 7 83 9 80 7 61 8* 54 5
165 min 115 12 115 7 66 9 67 9 81 10 80 7 60 8 57 6
All values are given as mean SD. Only time points selected for the statistical analysis (ANOVA) are shown.
*Signicant difference between placebo and BIBN4096BS pretreatment (P .005, paired t test).
Fig 1. Diameter (in millimeters) of supercial temporal artery and radial artery. Asterisk denotes
signicant difference (P .05) between placebo and BIBN4096BS pretreatment on human
-calcitonin generelated peptide (h-CGRP)induced vasodilatation (paired t test). Open squares,
Temporal artery, placebo pretreatment; open circles, radial artery, placebo pretreatment; closed
squares, temporal artery, 2.5-mg BIBN4096BS pretreatment; solid circles, radial artery, 2.5-mg
BIBN4096BS pretreatment (mean SD).
CLINICAL PHARMACOLOGY & THERAPEUTICS
208 Petersen et al MARCH 2005
C-bers surrounding cerebral and extracerebral arteries
and to the trigeminal ganglion cell bodies.
20
CGRP is present in plasma from healthy volun-
teers at rest.
2
We have previously shown that infu-
sion of the selective CGRP receptor antagonist
BIBN4096BS did not alter CBF or the diameter of
cerebral and peripheral arteries.
21
Hence circulating
levels of CGRP do not seem to exert a tonic dilator
action in these vascular beds. During a migraine
attack
5
and between attacks, CGRP is increased in
venous blood,
22
implicating a role of CGRP in mi-
graine pathogenesis. During an attack, the plasma
levels were increased (2-2.5 times) compared with
normal controls.
5
The infusion of 1.5 g/min of
h-CGRP in the current study increased the plasma
concentration approximately 3 to 4 times.
In healthy volunteers the infusion of CGRP has pre-
viously been shown to induce a sensation of fullness in
the head or mild headache,
23,24
corresponding to a
headache score of 1 in the current study. Because the
Fig 2. Plasma concentration of BIBN4096BS after 10-minute infusion of 2.5 mg. Values are given
as mean SD.
Fig 3. Plasma concentration of calcitonin generelated peptide (CGRP) on 2 different trial days.
Two asterisks denote a signicant difference (P .001) between BIBN4096BS and placebo
pretreatments (mean SD) (paired t test).
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):202-13 CGRP receptor antagonism and vascular headache 209
study was performed in healthy volunteers, a genuine
migraine attack was not expected. Why a migraine
attack is induced only in migraine patients and not in
healthy volunteers is most likely a threshold phenom-
enon. Through use of the human experimental head-
ache model, previously performed studies have shown
that headache response and MCA changes in healthy
volunteers compose a valid model for investigations of
migraine pathogenesis.
25-27
Methods of pharmacologic studies of cephalic he-
modynamics relevant to migraine. We have developed
the combination of methods applied in this study to
provide the most extensive and precise description of
pharmacologic effects on cephalic hemodynamics as
related to migraine. For CBF measurement, positron
emission tomography (PET) and magnetic resonance
(MR) techniques provide better spatial solution than
133
Xe SPECT, but this is of minor importance in phar-
macologic studies because no major variation has been
described in the response of cortical CBF to drugs.
28-30
Our pharmacologic experiments last several hours, and
the costs of blocking MR and especially PET equip-
ment for half a day or more on at least 30 experimental
days in 1 drug study are prohibitive in most centers.
Furthermore, studies of migraine patients in whom nau-
sea and vomiting may develop are very difcult inside
a magnet. The day-to-day CV of
133
Xe SPECT mea-
surements is 8%.
29
These gures are roughly equal to
those of PET and slightly better than MR-determined
CBF.
31
TCD measurements of velocity (V
mean
) as
stand-alone measurements are difcult to interpret be-
cause they depend both on the diameter of the MCA
and on CBF in the MCA territory (rCBF
MCA
), but they
are useful in combination with quantitative rCBF. TCD
measurements are most precise in the MCA, with a CV
of 26% between subjects, 20% between sides, 16%
between days, and 7% 5 minutes apart.
13
In studies of
headache, xed probes and continuous measurements
cannot be used because the probes cause local discom-
fort and headache.
32
Because of the relatively high
intersubject and interobserver variability of TCD mea-
surements, the current study was designed as a cross-
over study and the same trained investigator performed
all measurements. Power calculations based on an SD
of TCD measurements of 7.8 cm/s and allowing a 5%
type I and a 20% type II error showed that 7 subjects
were needed in a crossover design to detect a difference
of 15 cm/s (change from baseline of approximately
25%).
13
An alternative to these measurements could be MR
angiography, but MR angiography is difcult to time
with CBF measurements and poses problems in relation
to migraine induction. Its precision is not yet suf-
ciently documented and may not be high enough to
detect modest pharmacologic effects. In summary, the
combination of methods applied in this study remains
preferable to other alternatives for a broad character-
ization of the effects of messenger molecules and drugs
in the cephalic circulation.
Hemodynamic effects of CGRP. CGRP increases
CBF in dogs
33
and striatal blood ow in rats.
34
In
guinea pigs a pretreatment with urea was necessary
before CGRP-induced CBF changes could be in-
duced.
35
The peptide is partly responsible for cerebral
vasodilatation in response to hypotension,
36
postocclu-
sive hyperemia,
37
and cortical spreading depression.
38
In humans h-CGRP (0.6 g/min for 3 hours)
caused a signicant decrease of approximately
6 mL 100 g brain tissue
1
min
1
in the left hemi-
sphere CBF and a nonsignicant decrease of 5 mL
100 g brain tissue
1
min
1
in the right hemisphere in
one study.
39
The decrease was seen 30 minutes after the
initiation of the CGRP infusion but not after 2.5 hours
and was attributed to a decrease in PETCO
2
. Design, area
of measurement, and method of administration of
CGRP were different from those in this study, which
showed a slight but signicant increase in global and
regional CBF without changes in PETCO
2
.
In healthy volunteers (n 5) long-term infusion of
CGRP with a maximum dosage of 1.15 g/min caused
a nonsignicant decrease in the blood ow velocity in
the MCA (60 cm/s to 54 cm/s); CBF was not mea-
sured.
40
A similar decrease was seen with 0.6 g/
min.
39
In patients studied after subarachnoidal hemor-
rhage, V
MCA
decreased only on the side of the bleeding.
It was suggested that preconstriction of the artery or a
disruption of the blood-brain barrier was necessary for
CGRP to dilate MCA.
41,42
In the current study V
MCA
did not change after CGRP. We found, however, that
global CBF, as well as rCBF
MCA
, increased after CGRP
administration and calculated that CGRP increased the
diameter of the MCA
16
(Table III).
BIBN4096BS pretreatment did not prevent the
CGRP-induced increase in global CBF or rCBF
MCA
.
Furthermore, it did not prevent the CGRP-induced in-
crease in MCA diameter. However, the study was not
powered to rule out effects on MCA diameter, which is
calculated from 2 observed values and, therefore, is
more variable than the parameters alone.
In healthy volunteers and patients with subarachnoi-
dal hemorrhage, h-CGRP did not affect systolic or
diastolic BP. HR was increased signicantly.
24,41
In
healthy male volunteers 1.18 g/min induced a 30%
CLINICAL PHARMACOLOGY & THERAPEUTICS
210 Petersen et al MARCH 2005
increase in HR.
43,44
In the current study we found
similar effects of CGRP.
CGRP receptors and CGRP receptor blockade. The
CGRP receptor can be subdivided into 2 subtypes,
CGRP
1
and CGRP
2
; however, some discussion of the
validity of this classication exists.
45
The CGRP
1
re-
ceptor predominates in cerebral arteries, trigeminal
ganglion cell bodies, and perivascular nerve bers.
4,46
BIBN4096BS is the rst CGRP receptor antagonist
available for human clinical trials. BIBN4096BS has
high specicity and afnity for the human CGRP re-
ceptor; in human SK-N-MC (neuroblastoma cell line of
human origin) cells, it displayed an afnity of K
i
of
14.4 6.3 pmol/L, and for rat spleen, it demonstrated
a K
i
of 3.4 0.5 nmol/L.
47
The antagonist is 10 times
more potent in blocking rat or h-CGRP and h-CGRP
responses compared with the peptide antagonist
CGRP
8-37
. Furthermore, it was 10 times more potent in
inhibiting responses in rat atrium than in rat vas defer-
ens.
48
These results, together with the ndings that
SK-N-MC cell lines only express the CGRP
1
recep-
tor,
49
indicate that BIBN4096BS preferably binds to
this receptor subtype. BIBN4096BS inhibits both
CGRP-induced BP changes in rats and the neurogenic-
induced increase in facial blood ow in marmosets.
47,50
In isolated human middle cerebral, meningial, and tem-
poral arteries, BIBN4096BS blocks the dilatatory re-
sponse to CGRP.
9,51
In our study, the CGRP plasma concentration was
found to be signicantly higher on BIBN4096BS pre-
treatment days than on placebo pretreatment days. An
explanation would be a cross-reaction between
BIBN4096BS and CGRP, but this was ruled out. The
nding can be explained by a complete blockade of
CGRP receptors by the antagonist, resulting in a higher
concentration of circulating CGRP. This explanation
remains hypothetic until conrmed by further studies.
Our study showed that BIBN4096BS completely in-
hibits the effects of CGRP on the supercial temporal
and radial artery, as well as its effect on HR. In con-
trast, it had no signicant effect on the CGRP-mediated
CBF and V
MCA
increase. These results are important
with regard to the understanding of the action of CGRP
blockade in the treatment of acute migraine. In a phase
II proof-of-concept study in migraine patients,
BIBN4096BS showed a signicant effect in aborting
migraine attacks.
7
In agreement with this clinical result,
BIBN4096BS completely prevented CGRP-induced
headache in normal volunteers in our study.
In summary, our study has demonstrated the potency
of the CGRP receptor antagonist BIBN4096BS in hu-
mans. Thus it prevented all symptoms and signs elicited
by h-CGRP and caused no symptoms on its own.
More importantly, h-CGRPelicited headache was
completely prevented. Given that the compound inhib-
ited all effects of h-CGRP on extracerebral arteries
but had no effect on the induced increase in CBF or on
the calculated diameter of the MCA, it is suggested that
BIBN4096BS prevents or treats headache predomi-
nantly in an extracerebral manner. Whether the effect
takes place in the dura mater or in extracranial arteries
and whether areas of the brain stem or hypothalamus
devoid of a blood-brain barrier also play a role remain
to be determined.
We thank Lene Elkjr and Kirsten Bruunsgaard for excellent
technical support and Dr Kirsten Kasse for making everything work
perfectly.
None of the authors at the Danish Headache Center received
payment or honoraria. Lynna Lesko is an employee of Boehringer
Ingelheim Pharmaceuticals Inc. However, she did not receive special
honoraria from the company and has no equity or other ownership
interest.
References
1. Uddman R, Edvinsson L, Ekblad E, Hakanson R, Sundler
F. Calcitonin gene-related peptide (CGRP): perivascular
distribution and vasodilatory effects. Regul Pept 1986;
15:1-23.
2. Schifter S. Circulating concentrations of calcitonin gene-
related peptide (CGRP) in normal man determined with a
new, highly sensitive radioimmunoassay. Peptides 1991;
12:365-9.
3. Brain SD, Williams TJ, Tippins JR, Morris HR, MacIn-
tyre I. Calcitonin gene-related peptide is a potent vaso-
dilator. Nature 1985;313:54-6.
4. Jansen I, Mortensen A, Edvinsson L. Characterization of
calcitonin gene-related peptide receptors in human cere-
bral vessels. Vasomotor responses and cAMP accumula-
tion. Ann N Y Acad Sci 1992;657:435-40.
5. Goadsby PJ, Edvinsson L, Ekman R. Vasoactive peptide
release in the extracerebral circulation of humans during
migraine headache. Ann Neurol 1990;28:183-7.
6. Lassen LH, Haderslev PA, Jacobsen VB, Iversen HK,
Sperling B, Olesen J. CGRP may play a causative role in
migraine. Cephalalgia 2002;22:54-61.
7. Olesen J, Diener HC, Husstedt IW, Goadsby PJ, Hall D,
Meier U, Pollentier S, et al. Calcitonin gene-related pep-
tide receptor antagonist BIBN 4096 BS for the acute
treatment of migraine. N Engl J Med 2004;350:1104-10.
8. Schindler M, Doods HN. Binding properties of the novel,
non-peptide CGRP receptor antagonist radioligand
[(3)H]BIBN4096BS. Eur J Pharmacol 2002;442:187-93.
9. Verheggen R, Bumann K, Kaumann AJ. BIBN4096BS is
a potent competitive antagonist of the relaxant effects of
alpha-CGRP on human temporal artery: comparison with
CGRP(8-37). Br J Pharmacol 2002;136:120-6.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):202-13 CGRP receptor antagonism and vascular headache 211
10. Good Clinical Practice: consolidated guideline (E6) step 4
adopted 1996. Available from: URL: http://www.ich.org.
11. Classication and diagnostic criteria for headache disor-
ders, cranial neuralgias and facial pain. Headache Clas-
sication Committee of the International Headache So-
ciety. Cephalalgia 1988;8:1-96.
12. Kanno I, Lassen NA. Two methods for calculating re-
gional cerebral blood ow from emission computed to-
mography of inert gas concentrations. J Comput Assist
Tomogr 1979;3:71-6.
13. Thomsen LL, Iversen HK. Experimental and biological
variation of three-dimensional transcranial Doppler mea-
surements. J Appl Physiol 1993;75:2805-10.
14. Nielsen TH, Iversen HK, Tfelt-Hansen P. Determination
of the luminal diameter of the radial artery in man by
high frequency ultrasound: a methodological study. Ul-
trasound Med Biol 1990;16:787-91.
15. Shah VP, Midha KK, Dighe S, McGilveray IJ, Skelly JP,
Yacobi A, et al. Analytical methods validation: bioavail-
ability, bioequivalence and pharmacokinetic studies.
Conference report. Eur J Drug Metab Pharmacokinet
1991;16:249-55.
16. Dahl A, Russell D, Nyberg-Hansen R, Rootwelt K. Effect
of nitroglycerin on cerebral circulation measured by
transcranial Doppler and SPECT. Stroke 1989;20:
1733-6.
17. Sorteberg W. Cerebral artery velocity and cerebral blood
ow. In: Newell DW, Aaslid R, editors. Transcranial
Doppler. New York: Raven Press; 1992. p. 57-66.
18. Dahl A, Russell D, Nyberg-Hansen R, Rootwelt K. A
comparison of regional cerebral blood ow and middle
cerebral artery blood ow velocities: simultaneous mea-
surements in healthy subjects. J Cereb Blood Flow Metab
1992;12:1049-54.
19. Iovino M, Feifel U, Yong CL, Wolters JM, Wallenstein
G. Safety, tolerability and pharmacokinetics of BIBN
4096 BS, the rst selective small molecule calcitonin
gene-related peptide receptor antagonist, following single
intravenous administration in healthy volunteers. Ceph-
alalgia 2004;24:645-56.
20. Edvinsson L, Ekman R, Jansen I, Ottosson A, Uddman R.
Peptide-containing nerve bers in human cerebral arter-
ies: immunocytochemistry, radioimmunoassay, and in
vitro pharmacology. Ann Neurol 1987;21:431-7.
21. Petersen K, Lassen LH, Birk S, Olesen J. The novel
CGRP-antagonist BIBN4096BS does not affect the cere-
bral hemodynamics in healthy volunteers [abstract].
Cephalalgia 2003;23:729.
22. Ashina M, Bendtsen L, Jensen R, Schifter S, Olesen J.
Evidence for increased plasma levels of calcitonin gene-
related peptide in migraine outside of attacks. Pain 2000;
86:133-8.
23. Struthers AD, Brown MJ, Macdonald DW, Beacham JL,
Stevenson JC, Morris HR, et al. Human calcitonin gene
related peptide: a potent endogenous vasodilator in man.
Clin Sci 1986;70:389-93.
24. Howden CW, Logue C, Gavin K, Collie L, Rubin PC.
Haemodynamic effects of intravenous human calcitonin-
gene-related peptide in man. Clin Sci (Lond) 1988;74:
413-8.
25. Kruuse C, Thomsen LL, Jacobsen TB, Olesen J. The
phosphodiesterase 5 inhibitor sildenal has no effect on
cerebral blood ow or blood velocity, but nevertheless
induces headache in healthy subjects. J Cereb Blood
Flow Metab 2002;22:1124-31.
26. Kruuse C, Thomsen LL, Birk S, Olesen J. Migraine can
be induced by sildenal without changes in middle cere-
bral artery diameter. Brain 2003;126:241-7.
27. Thomsen L. Investigations into the role of nitric oxide
and the large intracranial arteries in migraine headache.
Cephalalgia 1997;17:873-95.
28. Edvinsson L, Krause DN. Cerebral blood ow and me-
tabolism. 2nd ed. Philadelphia: Lippincott Williams &
Wilkins; 2002.
29. Vorstrup S. Tomographic cerebral blood ow measure-
ments in patients with ischemic cerebrovascular disease
and evaluation of the vasodilator capacity by the acet-
azolamide test. Acta Neurol Scand Suppl 1988;114:1-48.
30. Vorstrup S, Henriksen L, Paulson OB. Effect of acetazol-
amide on cerebral blood ow and cerebral metabolic rate
for oxygen. J Clin Invest 1984;74:1634-9.
31. Carroll TJ, Teneggi V, Jobin M, Squassante L, Treyer V,
Hany TF, et al. Absolute quantication of cerebral blood
ow with magnetic resonance, reproducibility of the
method, and comparison with H2(15)O positron emission
tomography. J Cereb Blood Flow Metab 2002;22:1149-56.
32. Thomsen LL, Ahlgren M, Iversen HK, Olesen J. Middle
cerebral artery blood velocity increases during experi-
mental pressure pain in the head. In: Olesen J, Moskowitz
MA, editors. Experimental headache models. New York:
Raven Press; 1995. p. 337-9.
33. Baskaya MK, Suzuki Y, Anzai M, Seki Y, Saito K,
Takayasu M, et al. Effects of adrenomedullin, calcitonin
gene-related peptide, and amylin on cerebral circulation
in dogs. J Cereb Blood Flow Metab 1995;15:827-34.
34. Suzuki Y, Satoh S, Ikegaki I, Okada T, Shibuya M,
Sugita K, et al. Effects of neuropeptide Y and calcitonin
gene-related peptide on local cerebral blood ow in rat
striatum. J Cereb Blood Flow Metab 1989;9:268-70.
35. Beattie DT, McNeil DK, Connor HE. The inuence of
neurokinins and calcitonin gene-related peptide on cere-
bral blood ow in anaesthetized guinea-pigs. Neuropep-
tides 1993;24:343-9.
36. Hong KW, Pyo KM, Lee WS, Yu SS, Rhim BY. Phar-
macological evidence that calcitonin gene-related peptide
is implicated in cerebral autoregulation. Am J Physiol
1994;266:H11-6.
37. Macfarlane R, Moskowitz MA, Tasdemiroglu E, Wei EP,
Kontos HA. Postischemic cerebral blood ow and neu-
roeffector mechanisms. Blood Vessels 1991;28:46-51.
38. Wahl M, Schilling L, Parsons AA, Kaumann A. Involve-
ment of calcitonin gene-related peptide (CGRP) and nitric
CLINICAL PHARMACOLOGY & THERAPEUTICS
212 Petersen et al MARCH 2005
oxide (NO) in the pial artery dilatation elicited by cortical
spreading depression. Brain Res 1994;637:204-10.
39. Stanley J, Martin J, Richards H, Lovick M, Barron M,
Pickard J. Effects of calcitonin gene related peptide on
cerebral blood ow in healthy volunteers [abstract].
J Cereb Blood Flow Metab 1991;11(Suppl 2):s273.
40. Naylor AR, Miller JD, Edwards CR, Merrick MV, Sellar
RJ, OShaughnessy D, et al. Effect of calcitonin gene-
related peptide on middle cerebral artery velocities [ab-
stract]. J Cereb Blood Flow Metab 1989;9.
41. Juul R, Aakhus S, Bjornstad K, Gisvold SE, Brubakk AO,
Edvinsson L. Calcitonin gene-related peptide (human alpha-
CGRP) counteracts vasoconstriction in human subarach-
noid haemorrhage. Neurosci Lett 1994;170:67-70.
42. Naylor AR, Robertson IJ, Edwards CR, Merrick MV,
Sellar RJ, OShaughnessy D, et al. Cerebral vasospasm
following subarachnoid hemorrhage: effect of calcitonin
gene-related peptide on middle cerebral artery velocities
using transcranial Doppler sonography. Surg Neurol
1991;36:278-80.
43. Johnston FG, Bell BA, Robertson IJ, Miller JD, Haliburn
C, OShaughnessy D, et al. Effect of calcitonin-gene-
related peptide on postoperative neurological decits af-
ter subarachnoid haemorrhage. Lancet 1990;335:869-72.
44. Salmon P, Fitzgerald D, Lambe R, Darragh A,
OShaughnessy D, Riddell A, et al. A single rising intrave-
nous dose tolerance and pharmacodynamic study of human
calcitonin gene related peptide (CGRP) in healthy male
volunteers [abstract]. Clin Pharmacol Ther 1989;45:170.
45. Poyner DR, Sexton PM, Marshall I, Smith DM, Quirion
R, Born W, et al. International Union of Pharmacology.
XXXII. The mammalian calcitonin gene-related pep-
tides, adrenomedullin, amylin, and calcitonin receptors.
Pharmacol Rev 2002;54:233-46.
46. Edvinsson L, Cantera L, Jansen-Olesen I, Uddman R.
Expression of calcitonin gene-related peptide1 receptor
mRNA in human trigeminal ganglia and cerebral arteries.
Neurosci Lett 1997;229:209-11.
47. Doods H, Hallermayer G, Wu D, Entzeroth M, Rudolf K,
Engel W, et al. Pharmacological prole of BIBN
4096BS, the rst selective small molecule CGRP antag-
onist. Br J Pharmacol 2000;129:420-3.
48. Wu D, Eberlein W, Rudolf K, Engel W, Hallermayer G,
Doods H. Characterisation of calcitonin gene-related
peptide receptors in rat atrium and vas deferens: evidence
for a [Cys(Et)(2, 7)]hCGRP-preferring receptor. Eur
J Pharmacol 2000;400:313-9.
49. McLatchie LM, Fraser NJ, Main MJ, Wise A, Brown J,
Thompson N, et al. RAMPs regulate the transport and
ligand specicity of the calcitonin-receptor-like receptor.
Nature 1998;393:333-9.
50. Doods H, Rudolf K, Hallermayer G, Eberlein W. CGRP
antagonistic properties of BIBN 4096BS in rats. Cepha-
lalgia 2000;20:414-20.
51. Edvinsson L, Alm R, Shaw D, Rutledge RZ, Koblan KS,
Longmore J, et al. Effect of the CGRP receptor antago-
nist BIBN4096BS in human cerebral, coronary and
omental arteries and in SK-N-MC cells. Eur J Pharmacol
2002;434:49-53.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):202-13 CGRP receptor antagonism and vascular headache 213
CLINICAL TRIALS
Peginterferon alfa-2a does not alter the
pharmacokinetics of methadone in patients
with chronic hepatitis C undergoing
methadone maintenance therapy
Objective: Our objective was to quantify the pharmacokinetics of methadone and the pharmacokinetics and
pharmacodynamics of peginterferon alfa-2a (40 kd) in patients with chronic hepatitis C undergoing meth-
adone maintenance therapy.
Methods: Adults with chronic hepatitis C who had been receiving a consistent methadone maintenance
regimen for at least 3 months were eligible for this open-label, multicenter, nonrandomized drug interaction
study. All patients received 180 g subcutaneous peginterferon alfa-2a once weekly for 4 weeks and contin-
ued their methadone regimen. Serial blood samples were collected at baseline and immediately before and for
up to 168 hours after study drug administration for the purposes of quantifying methadone and peginter-
feron alfa-2a serum concentrations, measuring serum 2,5-oligoadenylate synthetase activity, and determin-
ing hepatitis C virus ribonucleic acid levels.
Results: Twenty-four patients were enrolled. Methadone exposure, as measured by maximum serum concen-
tration (C
max
) and area under the concentration-time curve (AUC) normalized to a 100-mg/d dose, after 4
doses of peginterferon alfa-2a increased by 10% to 15% when compared with baseline. The week 4/baseline
ratio of the mean C
max
was 1.11 (90% confidence interval [CI], 1.02-1.22), and for AUC from time 0 to 24
hours, the week 4/baseline ratio was 1.15 (90% CI, 1.08-1.23). The mean accumulation ratios (week 4/first
dose) for C
max
and AUC from time 0 to 168 hours of peginterferon alfa-2a were 2.1 and 2.3, respectively.
Conclusions: Peginterferon alfa-2a does not appreciably alter the pharmacokinetics of methadone. (Clin
Pharmacol Ther 2005;77:214-24.)
Mark Sulkowski, MD, Teresa Wright, MD, Stephen Rossi, PharmD,
Sanjeev Arora, MD, Matthew Lamb, PharmD, Ka Wang, PhD,
Jean-Michel Gries, PhD, and Sreeni Yalamanchili, PharmD Baltimore, Md, San Francisco and
Thousand Oaks, Calif, Albuquerque, NM, and Nutley, NJ
Injection drug use is the most common mode of
acquisition of new hepatitis C virus (HCV) infections in
the United States.
1,2
Injection drug users (IDUs) are at
high risk of HCV infection because of the efciency
with which the virus is parenterally transmitted. Indeed,
within 1 year of needle use, approximately 77% of
intravenous drug users become infected with HCV
3
; in
those who have used injectable drugs for 10 years or
more, the prevalence of HCV exposure is as high as
94%.
4,5
Comprehensive strategies for the management
From the Viral Hepatitis Center, Johns Hopkins University School
of Medicine, Baltimore; Gastrointestinal Section, Veterans
Affairs Medical Center, University of California, San Fran-
cisco, San Francisco; Science Center, University of New Mex-
ico, Albuquerque; Roche, Nutley; and Amgen, Inc, Thousand
Oaks.
The study was sponsored by Roche, Nutley, NJ.
Received for publication June 9, 2004; accepted Sept 16, 2004.
Reprint requests: Mark Sulkowski, MD, Viral Hepatitis Center, Johns
Hopkins Medical Institutions, 1830 E Monument St, Room 448,
Baltimore, MD 21205.
E-mail: msulkows@jhmi.edu
0009-9236/$30.00
Copyright 2005 by the American Society for Clinical Pharmacology
and Therapeutics.
doi:10.1016/j.clpt.2004.09.008
214
of HCV infection must encompass IDUs. Methadone
maintenance programs are effective in diminishing opi-
oid drug use and decreasing other high-risk behaviors;
accordingly, such drug treatment programs may serve
as an important link to provide IDUs with access to
treatment for chronic hepatitis C.
6
The combination of pegylated interferon plus ribavi-
rin, the treatment of choice in patients with chronic
hepatitis C,
7
produces sustained virologic responses in
52% to 63% of patients with HCV monoinfection and
40% of patients with HCV/human immunodeciency
virus coinfection.
8-12
Until recently, methadone treat-
ment has been a strict exclusion criterion in clinical
trials, and as a result, patients with hepatitis C who are
receiving methadone maintenance therapy have been
identied as an understudied population.
13
Thus this
study was designed to address this critical problem of
chronic hepatitis C in methadone recipients.
Because of the magnitude of the epidemic of hepa-
titis C in IDUs, it is important to understand the clinical
implications of the concomitant use of opiates (ie,
methadone) and anti-HCV therapies. Moreover, poten-
tial pharmacokinetic interactions between therapies for
hepatitis C and opiate agonists need to be investigated
to establish the safety of concurrent HCV and metha-
done treatment.
6
Methadone is a racemic mixture of which the
R-enantiomer is the active moiety.
14-17
There is marked
interindividual variability in the distribution and clear-
ance of the drug that is derived from variability in the
individual serum concentrations of
1
-acid glycopro-
tein and cytochrome P450 (CYP) 3A4 activity.
14,16,17
The stereoselective metabolism of methadone is medi-
ated by CYP3A4, CYP2C8, and CYP2D6, and data
obtained in vitro and in vivo demonstrate that there is
signicant potential for drug interactions with inhibi-
tors of these CYP isozymes.
18,19
Conventional interferon alfa has been shown to in-
hibit various CYP isozymes including CYP3A4 and
CYP2D6.
20-23
Moreover, some patients with chronic
hepatitis C have antibodies directed at CYP2D6,
24,25
and a study conducted during the rst month of treat-
ment with interferon alfa revealed an increase in the
activity of CYP3A4 and CYP2D6 in patients with
chronic hepatitis C.
26
It is, therefore, important to as-
sess the potential for clinically signicant CYP-
mediated interactions in drugs that may be used by
patients with chronic hepatitis C who are candidates for
interferon-based therapy.
Methadone has been shown to have anti-interferon
properties in mice.
27
Whether the reduction in serum
levels was related to decreased production or increased
clearance of endogenous interferon is another unre-
solved but relevant question with potential implications
for the treatment of chronic hepatitis C in patients
undergoing methadone maintenance therapy.
The objective of this study was to quantitatively
evaluate the effects of peginterferon alfa-2a on the
pharmacokinetics of methadone in patients with
chronic hepatitis C who were receiving a stable meth-
adone maintenance regimen. The pharmacokinetics and
pharmacodynamics of peginterferon alfa-2a were also
quantied in these individuals.
METHODS
Subjects. Eligible subjects were aged 18 years or
greater; had evidence of HCV infection, dened as a
positive anti-HCV antibody test result and either a
positive radioimmunoblot assay for HCV or quanti-
able HCV ribonucleic acid (RNA) in serum by poly-
merase chain reaction (COBAS AMPLICOR HCV
MONITOR Test, v2.0; Roche Diagnostics, Basel, Swit-
zerland) (limit of detection, 2000 IU/mL); and had
compensated liver disease. Women of childbearing po-
tential were required to have a negative urine preg-
nancy test result documented within 24 hours before
administration of the rst dose of peginterferon alfa-2a.
All participants had to have been receiving a consis-
tent methadone maintenance regimen, in terms of the
formulation, dose, and frequency of administration, for
at least 3 months before study entry. No specic dose of
methadone was selected for the trial. Because alcohol
inuences the disposition of methadone,
28
patients
were not eligible if the dosage form of methadone
included alcohol (ie, use of methadone tincture or
United States Pharmacopeia oral solution containing
8% alcohol was prohibited). Methadone diskettes, tab-
lets, alcohol-free liquid concentrate, and powder were
acceptable dosage forms.
Patients were excluded if they had a positive urine
screening result for drugs, except methadone, or evi-
dence of excessive alcohol or drug use within 6 months
of study entry. Drug use was dened as the use of drugs
without a documented indication, including amphet-
amines, barbiturates, benzodiazepines, cannabis, co-
caine, or opiates other than methadone. A history of
severe psychiatric disease, including depression, was
also grounds for exclusion.
Patients coinfected with hepatitis A or B virus or
human immunodeciency virus were also excluded, as
were patients with evidence of severe or advanced liver
disease of any etiology or other severe or chronic
systemic disease.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):214-24 Peginterferon alfa-2a and methadone PK 215
Prescription medications were allowed for the treat-
ment of concurrent chronic conditions, provided that
the patient had been receiving a consistent regimen for
at least 2 months and adjustments to the regimen were
not anticipated during the study and that the medication
was not a known inhibitor of CYP3A4 (ie, clarithro-
mycin, erythromycin, uconazole, uoxetine, uvox-
amine, itraconazole, ketoconazole, nefazodone, and
omeprazole). Because methadone is a substrate of
CYP3A4, consumption of grapefruit juice was prohib-
ited during the study. In addition to the above-
mentioned medications, the use of theophylline was
prohibited. With the exception of acetaminophen (INN,
paracetamol), which could be used at a maximum dose
of 4 g/d, the use of nonprescription medications was
prohibited for 2 weeks before and during the study.
Female participants were required to use contracep-
tion throughout the study.
Study design. Patients enrolled in this open-label,
multicenter, nonrandomized drug interaction study con-
tinued their ongoing daily methadone maintenance
therapy regimen and received 180 g subcutaneous
peginterferon alfa-2a (40 kd) (Pegasys; Roche, Nutley,
NJ) once weekly for 4 weeks. Patients fasted for 12
hours before peginterferon alfa-2a administration dur-
ing weeks 1 and 4. Caffeinated drinks were not allowed
for 48 hours before or on study drug administration
days. Patients remained in the study clinic for 24 hours
after peginterferon alfa-2a administration for blood
sampling. The date and time of methadone administra-
tion were recorded at baseline, on day 1 (week 1), and
on day 22 (week 4). The date and time of administra-
tion of all peginterferon alfa-2a doses and collection of
all blood samples were recorded.
Serial blood samples of 5 mL were collected at
baseline (day 7) and during week 4 (day 22) at 0 hours
and at 0.5, 1, 2, 3, 4, 6, 8, 12, and 24 hours after
methadone administration for the purpose of quantitat-
ing methadone serum levels.
Serial blood samples of 5 mL were collected during
week 1 (day 1) and week 4 (day 22) before adminis-
tration of peginterferon alfa-2a (0 hours) and at 3, 8, 24,
48, 72, 84, 96, 120, 144, and 168 hours after adminis-
tration for the purpose of quantitating peginterferon
alfa-2a serum levels.
Blood samples (2 mL) were collected before meth-
adone administration at baseline (day 7); before pegin-
terferon alfa-2a administration during weeks 1, 2, 3,
and 4; and before dosing and at 6, 12, 24, 36, 48, 72, 96,
and 168 hours after administration of peginterferon
alfa-2a during week 1 for the purpose of quantifying
serum HCV RNA levels. Serial blood samples were
also collected before dosing and at 3, 8, 24, 48, 72, 84,
96, 120, 144, and 168 hours after administration of
peginterferon alfa-2a at baseline (day 7) and during
weeks 1 and 4 for the purpose of measuring serum
2,5-oligoadenylate synthetase (2,5-OAS) activity, a
widely used marker for interferon-induced antiviral ac-
tivity.
Serum concentrations of methadone were quantied
by a validated gas chromatographymass spectroscopy
assay at PPD Development (Richmond, Va). Samples
were kept frozen until analyzed, at which time they
were thawed at room temperature in a water bath. One
milliliter of sample and 100 L of internal standard (50
ng/mL methadone-d
9
) were placed in a silylated 16
125mm culture tube. After the mixture was allowed to
stand for 20 minutes, 4 mL of water and 2 mL of
100-mmol/L phosphate buffer/methanol (80:20) were
added; the tube was blended in a vortex mixer and then
centrifuged at 3000 rpm for 5 minutes. The supernatant
was decanted into a second silylated 16 125mm
culture tube. The sample was then placed on a solid
phase extraction cartridge (Bond Elut Certify SPE col-
umns, 130 mg, LRC [large reservoir cartridge], Varian
No. 1211-3050; Varian, Inc, Palo Alto, Calif ). The
cartridge was then washed with 2 mL of water, 2 mL of
100-mmol/L acetic acid, and then 3 mL 100% metha-
nol. The columns were dried under vacuum for 5 min-
utes before being eluted into silylated, conical centri-
fuge tubes with 3 mL of dichloromethane/isopropyl
alcohol/ammonium hydroxide (78:20:2). The eluate
was then evaporated to dryness at 40C or greater under
a gentle stream of nitrogen. Samples were reconstituted
with 25 L of 10% N-methyl-N-trimethylsilyltriu-
oroacetamide in toluene. Reconstituted samples were
stored at 20C until analysis. A 1.0-L volume was
then injected onto a DB-1 30-m 0.32-mm inner
diameter column (J&W Scientic, Folsom, Calif)
mounted in a Varian 3400 gas chromatograph (Varian,
Inc) coupled with a mass spectrometer. The carrier gas
was helium, the injector temperature was 300C, and
the instrument was programmed to increase the column
temperature from 165C to 300C at a rate of 20C/
min. The analytic range of the assay was 0.1 to 50
ng/mL. Samples with concentrations greater than 50
ng/mL were diluted and reanalyzed. The coefcient of
variation for the assay ranged from 2.5% to 35.1%.
Serum concentrations of peginterferon alfa-2a were
quantied by a validated enzyme-linked immunosor-
bent assay at MDS Pharma Services (Montreal, Que-
bec, Canada). In a 1-step immunoreaction, the pegin-
terferon alfa-2a (40 kd) contained in the sample is
bound by the peroxidase-conjugated antibody. The as-
CLINICAL PHARMACOLOGY & THERAPEUTICS
216 Sulkowski et al MARCH 2005
say is a quantitative sandwich enzyme immunoassay
which uses 2 monoclonal antibodies that recognize
different epitopes. It is not known whether the antibod-
ies bind to receptor-bound peginterferon alfa-2a (40
kd). One hundred microliters of substrate buffer and 50
L of assay dilution buffer were added to all wells in
microtiter plates. A 50-L aliquot of the standard or
study sample was then added, followed by 50 L of
peroxidase-conjugated solution. This complex binds
via the captureantipeginterferon alfa-2a (40 kd) an-
tibody to the multiprotein layercoated surface of the
microtiter plate. The plate is sealed and incubated at
room temperature for 30 minutes with shaking at 150
rpm before incubation for 16 to 24 hours at 37C
without shaking. After incubation, the plate was shaken
at room temperature at 150 rpm for 60 minutes. The
plate was then washed 5 times with 300 L of wash
buffer. Next, 200 L of substrate solution (0.8 mL of
0.2% tetramethylbenzidine in acetone/ethanol/hydro-
gen peroxide [10:89:1]/22 mL substrate buffer, pH 4.1)
was added to each well. This solution is converted by
the peroxidase to a colored product that can be deter-
mined photometrically. The plates were sealed and
incubated in the dark at room temperature for approx-
imately 15 to 20 minutes with shaking at 150 rpm. Fifty
microliters of 2N sulfuric acid was then added to stop
the reaction, and the plate was read within 15 minutes
by use of a SpectraMax 340 PC microplate reader
(Molecular Devices Corporation, Sunnyvale, Calif) set
to 450 to 650 nm. The concentration of the samples was
calculated by comparison with a standard curve. The
analytic range of the assay was 350 to 3000 pg/mL.
Samples with concentrations greater than 3000 pg/mL
were diluted and reanalyzed. The coefcient of varia-
tion for the assay ranged from 9.1% to 13.5%.
HCV RNA levels in serum were determined with the
COBAS AMPLICOR HCV MONITOR Test, v2.0
(Roche Diagnostics, Basel, Switzerland). Serum 2,5-
OAS activity was determined with a commercially
available radioimmunoassay kit.
After completion of the study, patients were allowed
to participate in an open-label continuation study with
peginterferon alfa-2a and ribavirin combination ther-
apy. Patients who declined combination therapy were
followed up for 4 weeks after administration of the last
dose of peginterferon alfa-2a.
The institutional review boards at each center ap-
proved the protocol. All patients provided informed
written consent before enrollment. The study was con-
ducted in conformance with the principles of the Dec-
laration of Helsinki, the principles outlined in the
Guideline for Good Clinical Practice, and the laws and
regulations of the United States.
Pharmacokinetic parameters. The following phar-
macokinetic parameters were determined for peginter-
feron alfa-2a and methadone by use of noncompart-
mental methods: observed maximum serum drug
concentration (C
max
), time to C
max
(t
max
), time of last
observed serum concentration (t
last
), area under the
serum concentrationtime curve (AUC) (AUC between
time 0 and the time of the last observed serum concen-
tration [AUC
0-last
], AUC from time 0 extrapolated to
innity [AUC
0-
], and AUC from time 0 to 24 hours
[AUC
0-24
] for methadone and AUC from time 0 to 168
hours [AUC
0-168
] for peginterferon alfa-2a). The total
apparent clearance (CL
ss
/F) and apparent terminal
phase volume of distribution (V
z
/F) were calculated for
methadone, and the accumulation ratio for week 4:1 for
C
max
and AUC
0-168
was calculated for peginterferon
alfa-2a.
Pharmacokinetic characteristics of peginterferon
alfa-2a and methadone were evaluated by model-
independent analysis by use of WinNonlin Pro (version
4.0; Pharsight Corporation, Mountain View, Calif). Ac-
tual sampling times rather than scheduled sampling
times were used in all analyses. For individual pegin-
terferon alfa-2a proles with measurable baseline con-
centrations, all data for the subject were excluded if the
baseline concentration exceeded 1000 pg/mL. If the
baseline value was less than 1000 pg/mL, subsequent
values were adjusted for the baseline concentration by
subtracting the baseline value from the measured value.
After adjustment, if the value was below the limit of
quantitation, it was excluded from the analysis.
The slope of the terminal log-linear portion of each
peginterferon alfa-2a and methadone concentration-
time prole was determined by least squares linear
regression. The log-linear trapezoidal method with ex-
trapolation beyond the last experimental concentration
value was used to determine AUC
0-
.
The total apparent clearance of methadone (CL
ss
/F)
was calculated as Dose/AUC
0-24
; the apparent volume
of distribution of methadone (V
z
/F) was estimated as
Dose/(Terminal slope AUC
0-
).
Dose-normalized methadone pharmacokinetic pa-
rameters were used to compare the C
max
and AUC of
methadone at baseline and during week 4; the param-
eters were normalized to a 100-mg dose.
Pharmacodynamic parameters. 2,5-OAS is an im-
portant effector of interferon-induced antiviral effects,
and the serum activity of this enzyme is widely used as
an indicator of interferon activity. Calculations of the
following parameters were based on week-1 measure-
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):214-24 Peginterferon alfa-2a and methadone PK 217
ments of 2,5-OAS data: observed maximum activity
(OAS
max
), time to OAS
max
(t
max
), observed last activity
(OAS
last
), time to OAS
last
(t
last
), and area beneath the
effect curve at the last time point (ABEC
last
).
Serum 2,5-OAS activitytime proles were ana-
lyzed with noncompartmental methods by use of Win-
Nonlin Pro (version 4.0). The log-linear trapezoidal
method was used to determine ABEC
last
. For individual
proles with measurable baseline 2,5-OAS activity,
subsequent values were adjusted for the baseline con-
centration by subtracting the baseline value from the
measured value. After adjustment, if the value was
below the limit of quantitation, it was excluded from
the analysis.
Statistical analysis. No formal sample size calcula-
tion was made. A planned enrollment of 24 patients
was established, with the expectation that 18 individu-
als would complete the study.
A 90% condence interval (CI) for the true ratio of
week 4 to baseline for the C
max
and AUC
0-24
of meth-
adone and a 90% CI for the true difference between
means for the t
max
of methadone were calculated. The
residual mean square error from an ANOVA was used
as the estimate of variance for the CI.
RESULTS
A total of 24 patients were enrolled in the study, 22
of whom received 4 doses of peginterferon alfa-2a and
completed the pharmacokinetic portion of the study.
Data from 2 patients who received only 3 doses of
peginterferon alfa-2a were excluded from the pharma-
cokinetic analyses. In addition, pharmacokinetic data
for peginterferon alfa-2a from 2 further patients were
excluded because of analytic artifacts. All patients were
included in the safety analysis. The baseline character-
istics of the patients are presented in Table I. The mean
methadone dose at baseline was 88 mg/d.
Pharmacokinetic results
Inuence of peginterferon alfa-2a on pharmacoki-
netics of methadone. Methadone exposure after 4
doses of peginterferon alfa-2a increased by 10% to 15%
when compared with baseline (Table II). However, the
90% CIs for the ratio of the mean C
max
and AUC
0-24
at
baseline and during week 4 were 1.02 to 1.22 and 1.08
to 1.23, respectively, which fell within the generally
accepted equivalence interval of 0.80 to 1.25, indicat-
ing no signicant effect of peginterferon alfa-2a on the
pharmacokinetics of methadone (Table II).
Paired values for C
max
and AUC
0-24
for methadone
are presented for each patient in Fig 1, A and B,
respectively. Intersubject variabilities at baseline and
during week 4 were 36% and 34%, respectively, for
C
max
and 28% and 33%, respectively, for AUC
0-24
. Of
the patients, 8 of 22 (36%) and 16 of 21 (76%) had
higher individual C
max
and AUC
0-24
values, respec-
tively, during week 4 than at baseline.
The mean 24-hour serum concentration prole for
methadone (normalized to a 100-mg dose) at baseline
and during week 4 is presented in Fig 2. Methadone
Fig 1. Individual dose-normalized maximum serum concen-
tration (C
max
) (A) and area under concentration-time curve
from time 0 to 24 hours (AUC
0-24
) (B) for methadone at
baseline and during week 4.
Table I. Baseline characteristics of 24 patients
enrolled in study
Characteristic No.
Male/female 15/9
Race (white/black/other) 13/6/5
Age (y) (median and range) 49.5 (33-61)
Weight (kg) (median and range) 78.5 (52-129)
Height (cm) (median and range) 170 (137-185)
Body mass index (kg/m
2
) (median
and range)
26.9 (19.1-48.9)
Methadone dose (mg/d) (median
and range)
95 (30-150)
CLINICAL PHARMACOLOGY & THERAPEUTICS
218 Sulkowski et al MARCH 2005
concentrations were higher at each time point during
the 24-hour sampling period.
All patients continued to receive stable doses of
methadone throughout the study.
Inuence of methadone on pharmacokinetics of
peginterferon alfa-2a. Three patients had detectable
predose concentrations of peginterferon alfa-2a, and
subsequent determinations were adjusted before analy-
sis. The intersubject variability of C
max
was 38% and
that of AUC
0-168
was 39% after the rst dose of pegin-
terferon alfa-2a, and it was 42% for both parameters
during week 4.
Single- and multiple-dose pharmacokinetic data for
peginterferon alfa-2a in patients receiving methadone
maintenance therapy are presented in Table III. The
mean accumulation ratios (week 4/rst dose) for C
max
and AUC
0-168
were 2.1 (90% CI, 1.62-2.58) and 2.3
(90% CI, 1.82-2.78), respectively. Values for peginter-
feron alfa-2a during week 1 and week 4 reveal a general
increase during the study (Fig 3).
Pharmacodynamic results
Mean 2,5-OAS activity at baseline and after the
rst (week 1) and last doses of peginterferon alfa-2a
(week 4) are presented in Fig 4. A summary of 2,5-
OAS activity parameters is presented in Table IV. The
induction of 2,5-OAS activity in methadone recipi-
ents was similar to that previously reported in subjects
not receiving methadone maintenance therapy.
Of 24 patients, 12 (50%) had either a greater than
2-log
10
decrease in serum HCV RNA levels or unde-
tectable serum HCV RNA levels after 4 weeks of
treatment with peginterferon alfa-2a. Of these 12
Table II. Methadone noncompartmental pharmacokinetic parameters
Parameter Baseline (n 22) Week 4 (n 22)
t
max
(h) 2.9 1.7 (2.2-3.6) 2.6 1.0 (2.2-3.0)
C
max
(ng/mL)* 701 253 (595-807) 774 266 (663-885)
AUC
last
(ng h/mL)* 10,671 2974 (9428-11,914) 12,511 4300 (10,714-14,308)
AUC
0-24
(ng h/mL)* 10,576 3012 (9288-11,864) 12,576 4153 (10,841-14,311)
CL
ss
/F (L/h) 11 5 (8.9-12.2) 9 4 (7.3-10.7)
V
z
/F (L) 634 541 (397-871) 616 574 (376-856)
Geometric least squares means
C
max
(ng/mL) 656.7 731.3
Week 4/baseline (mean and 90% CI) 1.11 (1.02-1.22)
AUC
0-24
(ng h/mL) 10,063.9 11,602.3
Week 4/baseline (mean and 90% CI) 1.15 (1.08-1.23)
Values are given as mean SD with 95% CI, unless otherwise indicated.
t
max
, Time to maximum serum concentration; C
max
, maximum serum concentration; AUC
last
, AUC between time 0 and the time of the last observed serum
concentration; AUC
0-24
, area under concentration-time curve from time 0 to 24 hours; CL
ss
/F, total apparent clearance; V
z
/F, apparent volume of distribution; CI,
condence interval.
*Individual parameter estimates were normalized to a 100-mg dose.
n 21.
n 20.
Table III. Single- and multiple-dose peginterferon alfa-2a noncompartmental pharmacokinetic parameters in
patients receiving stable methadone maintenance therapy
Parameter First dose (n 22) Week 4 (n 20)
t
max
(h) 84 34 (70-98) 78 24 (67-89)
C
max
(ng/mL) 14.4 5.4 (12.1-16.7) 26.0 10.9 (21.2-30.8)
t
last
(h) 167 5 (164.9-169.1) 168 0.5 (167.8-168.2)
C
last
(ng/mL) 9.9 4.5 (8.0-11.8) 15.9 6.3 (13.1-18.7)
AUC
last
(ng h/mL) 1769 669 (1489-2049) 3459 1451 (2823-4095)
AUC
0-168
(ng h/mL) 1788 700 (1495-2081) 3515 1463 (2874-4156)
Accumulation ratio: C
max
NA 2.1 1.3
Accumulation ratio: AUC
0-168
NA 2.3 1.3
Values are given as mean SD with 95% CI.
t
last
, Time of last observed serum concentration; C
last
, serum concentration from time 0 to t
last
; AUC
0-168
, area under concentration-time curve from time 0 to 168 hours;
NA, not applicable.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):214-24 Peginterferon alfa-2a and methadone PK 219
week-4 responders, 8 had HCV RNA levels below the
limit of quantitation of the assay by week 4.
Safety
A total of 123 adverse events were reported by 20
patients (83%). The most common adverse events were
those typically associated with conventional interferon
or pegylated interferon administration (ie, ulike symp-
toms). Of the patients, 4 (17%) had a neutrophil count
of less than 0.75 10
9
/L at some point during the trial
but no patient had a neutrophil count of less than 0.50
10
9
/L. Of the patients, 2 (8%) had a platelet count of
Fig 2. Mean dose-normalized methadone serum concentration (Conc) (SD) at baseline and during
week 4 after 4 180-g doses of peginterferon alfa-2a.
Fig 3. Mean serum concentration of peginterferon alfa-2a (SD) during the rst and last (week 4)
dose intervals.
CLINICAL PHARMACOLOGY & THERAPEUTICS
220 Sulkowski et al MARCH 2005
less than 50 10
9
/L at some point during the trial but
no patient had a platelet count of less than 20 10
9
/L.
There were no serious adverse events, and no patients
withdrew prematurely because of adverse events or
laboratory abnormalities. In addition, no subject had
clinical signs or symptoms of intoxication or with-
drawal related to methadone.
DISCUSSION
The results of our study demonstrate that peginter-
feron alfa-2a does not inuence the pharmacokinetics
of methadone to a clinically signicant extent in pa-
tients receiving ongoing methadone maintenance ther-
apy. Baseline pharmacokinetic and pharmacodynamic
data for peginterferon alfa-2a in the absence of meth-
adone administration were not available for the patients
enrolled in our study; however, the pharmacokinetic
prole of peginterferon alfa-2a in these individuals was
generally similar to that observed previously in healthy
volunteers and patients with chronic hepatitis C.
29-32
The pharmacokinetic values for racemic methadone
reported in our study (normalized to 100 mg) compare
well with those reported previously in opiate users
(normalized to 10 mg
14
or 70 mg
16
). It must be noted
that we did not evaluate the pharmacokinetic properties
of the active R-isomer of methadone, so it is not pos-
sible to comment on the impact of peginterferon alfa-2a
on the pharmacokinetics of this entity. We also did not
monitor
1
-acid glycoprotein levels, which have an
inuence on the plasma pharmacokinetics of the drug.
15
As anticipated, the mean serum concentrations of
peginterferon alfa-2a were higher during week 4 than
during week 1. This agent has a long elimination half-
life and, as a result, does not reach steady-state levels
until 6 to 8 weeks after the initiation of weekly admin-
istration. The accumulation ratios (steady-state concen-
Table IV. Summary of baseline-adjusted
2,5-oligoadenylate synthetase activity at week 1 (N
24)
Parameter Mean
t
max
(h) 98 49 (78-118)
2,5-OAS
max
(nmol/L h
1
) 2.9 1.5 (2.3-3.5)
t
last
(h) 162 13 (157-167)
2,5-OAS
last
(nmol/L h
1
) 2.1 1.2 (1.6-2.6)
ABEC
last
(nmol/L) 253 139 (197-309)
Values are given as mean SD with 95% CI.
2,5-OAS
max
, Observed maximum 2,5-oligoadenylate synthetase activity;
2,5-OAS
last
, observed last 2,5-oligoadenylate synthetase activity; ABEC
last
,
area beneath effect curve at last time point.
Fig 4. Mean serum 2,5-oligoadenylate synthetase (OAS) activity-time proles (SD) at baseline,
during week 1, and during week 4.
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):214-24 Peginterferon alfa-2a and methadone PK 221
tration/maximum concentration after 1 dose) reported
previously have been approximately 2 to 3.
29,32
Thus
our nding of an accumulation ratio of approximately 2
after 4 weeks of weekly dosing is in line with our
expectations.
The biologic response to peginterferon alfa-2a, as
assessed by 2 5-OAS activity, was similar in patients
receiving ongoing methadone maintenance therapy in
our study and in healthy subjects enrolled in previous
studies.
31,33
The decrease in HCV RNA levels during
the rst 4 weeks of treatment with peginterferon alfa-2a
was typical of that seen in chronic hepatitis C patients
who are not receiving methadone maintenance therapy,
and one half of the patients had a substantial reduction
in HCV RNA after only 4 doses of peginterferon alfa-
2a.
34
Peginterferon alfa-2a monotherapy was well toler-
ated during methadone maintenance therapy in patients
with chronic hepatitis C, and no serious adverse events
were observed. The spectrum and frequency of adverse
events were similar to those reported in patients with
chronic hepatitis C who were not receiving methadone
in a phase III trial,
35
although it must be acknowledged
that the patients in our trial were monitored for only 4
weeks.
There are several limitations to our study. As noted,
we evaluated patients during an abbreviated course of
treatment with peginterferon alfa-2a monotherapy
rather than for the full 48-week duration recommended
for the treatment of chronic hepatitis C at the time the
study was planned. However, the 4-week treatment
period was sufcient to conrm whether peginterferon
alfa-2a interfered with the pharmacokinetics of metha-
done and to determine whether the pharmacodynamics
of peginterferon alfa-2a was impaired by methadone.
Thus, although the most important therapeutic end
point in patients with chronic hepatitis C is the eradi-
cation of HCV infection, our study was not designed
with sufcient statistical power to examine sustained
virologic response rates. In addition, it must be noted
that therapy for chronic hepatitis C has continued to
evolve since our study was conceived, and the current
treatment of choice is the combination of pegylated
interferon plus ribavirin.
7
Thus it remains to be deter-
mined whether ribavirin has any inuence on the phar-
macokinetics of methadone.
These limitations notwithstanding, our data provide
important information to clinicians treating the rela-
tively large population of HCV-infected persons receiv-
ing methadone maintenance therapy worldwide. De-
spite the size of the population, patients with chronic
hepatitis C who are receiving methadone maintenance
therapy have been understudied, a problem that was
recognized by the US National Institutes of Health
Consensus Panel in their 2002 recommendations for the
management of chronic hepatitis C.
7
However, there is
increasing interest in the treatment of hepatitis C in
methadone recipients, and recent studies have sug-
gested that sustained viral response rates in patients
receiving methadone maintenance therapy are similar
to those observed in other HCV-infected patient popu-
lations. For example, among patients included in a
large, retrospective, matched cohort study, Van Thiel et
al
36
reported that sustained virologic response rates
were similar in patients receiving methadone mainte-
nance therapy and in patients without a history of
intravenous drug use after treatment with daily conven-
tional interferon (33% and 37%, respectively). Interest-
ingly, the researchers observed that the methadone dose
was increased in some patients during treatment with
conventional interferon alfa. However, consistent with
our ndings, the authors reported that these increases in
methadone dose generally represented the use of meth-
adone to treat side effects of conventional interferon
rather than a response to clinical evidence of a phar-
macokinetic interaction. Similarly, in an interim anal-
ysis of 50 methadone maintenance patients treated with
conventional interferon plus ribavirin, Sylvestre
37
re-
ported that the end-of-treatment response rate was sim-
ilar to that of patients without a history of injection drug
use. It must be noted that we did not observe a similar
increase in methadone doses during our study. A
smaller, prospective, open-label German study has also
reported success in treating intravenous drug users en-
rolled in a detoxication program with conventional
interferon-based therapy.
38
Although preliminary, these
encouraging results suggest that patients with a recent
history of injection drug use, particularly those under-
going stable methadone maintenance therapy, can be
successfully treated for hepatitis C with interferon-
based regimens.
In conclusion, our data indicate that peginterferon
alfa-2a does not appreciably alter the pharmacokinetics
of methadone. Thus concurrent treatment with metha-
done is not a contraindication to therapy with peginter-
feron alfa-2a for chronic hepatitis C. Although the
decision to treat a patient who is using illicit drugs
should be made by the patient after consulting with his
or her physician, this decision need not be inuenced
by the use of methadone maintenance therapy, and a
priori adjustment of either the methadone or the pegin-
terferon alfa-2a dose is not recommended in patients
receiving stable methadone maintenance therapy who
CLINICAL PHARMACOLOGY & THERAPEUTICS
222 Sulkowski et al MARCH 2005
undergo treatment for chronic hepatitis C with pegin-
terferon alfa-2acontaining regimens.
Drs Sulkowski and Arora have received research grants from
Roche. Dr Wright has no potential conicts of interest. Dr Yalaman-
chili is a former employee of Roche. Drs Rossi, Lamb, Wang, and
Gries are currently employed by Roche.
References
1. Recommendations for prevention and control of hepatitis
C virus (HCV) infection and HCV-related chronic dis-
ease. Centers for Disease Control and Prevention.
MMWR Recomm Rep 1998;47:1-39.
2. Alter MJ. Prevention of spread of hepatitis C. Hepatology
2002;36:S93-8.
3. Garfein RS, Vlahov D, Galai N, Doherty MC, Nelson
KE. Viral infections in short-term injection drug users:
the prevalence of the hepatitis C, hepatitis B, human
immunodeciency, and human T-lymphotropic viruses.
Am J Public Health 1996;86:655-61.
4. Thomas DL. Correlates of hepatitis C virus infections
among injection drug users. Medicine (Baltimore) 1995;
74:212-20.
5. Novick DM. The impact of hepatitis C virus infection on
methadone maintenance treatment. Mt Sinai J Med 2000;
67:437-43.
6. Edlin BR. Prevention and treatment of hepatitis C in
injection drug users. Hepatology 2002;36:S210-9.
7. National Institutes of Health Consensus Development
Conference Statement: Management of hepatitis C:
2002June 10-12, 2002. Hepatology 2002;36:S3-20.
8. Manns MP, McHutchison JG, Gordon SC, Rustgi VK,
Shiffman M, Reindollar R, et al. Peginterferon alfa-2b
plus ribavirin compared with interferon alfa-2b plus riba-
virin for initial treatment of chronic hepatitis C: a ran-
domised trial. Lancet 2001;358:958-65.
9. Fried MW, Shiffman ML, Reddy KR, Smith C, Marinos
G, Goncales FL Jr, et al. Peginterferon alfa-2a plus
ribavirin for chronic hepatitis C virus infection. N Engl
J Med 2002;347:975-82.
10. Hadziyannis SJ, Sette H Jr, Morgan TR, Balan V, Diago
M, Marcellin P, et al. Peginterferon alfa-2a and ribavirin
combination therapy in chronic hepatitis C: randomized
study of treatment duration and ribavirin dose. Ann In-
tern Med 2004;140:346-55.
11. Zeuzem S, Diago M, Gane E, Reddy KR, Pockros P, Prati
D, et al. Peginterferon alfa-2a (40 kilodaltons) and ribavirin
in patients with chronic hepatitis C and normal aminotrans-
ferase levels. Gastroenterology 2004;127:1724-32.
12. Torriani FJ, Rodriguez-Torres M, Rockstroh JK, Lissen
E, Gonzalez-Garcia J, Lazzarin A, et al. Peginterferon
alfa-2a plus ribavirin for chronic hepatitis C virus infec-
tion in HIV-infected patients. N Engl J Med 2004;351:
438-50.
13. Strader DB. Understudied populations with hepatitis C.
Hepatology 2002;36:S226-36.
14. Wolff K, Rostami-Hodjegan A, Shires S, Hay AW, Feely
M, Calvert R, et al. The pharmacokinetics of methadone
in healthy subjects and opiate users. Br J Clin Pharmacol
1997;44:325-34.
15. Foster DJ, Somogyi AA, Dyer KR, White JM, Bochner
F. Steady-state pharmacokinetics of (R)- and (S)-
methadone in methadone maintenance patients. Br J Clin
Pharmacol 2000;50:427-40.
16. Foster DJ, Somogyi AA, White JM, Bochner F. Popula-
tion pharmacokinetics of (R)-, (S)- and rac-methadone in
methadone maintenance patients. Br J Clin Pharmacol
2004;57:742-55.
17. Boulton DW, Arnaud P, DeVane CL. Pharmacokinetics
and pharmacodynamics of methadone enantiomers after a
single oral dose of racemate. Clin Pharmacol Ther 2001;
70:48-57.
18. Wang JS, DeVane CL. Involvement of CYP3A4, CYP2C8,
and CYP2D6 in the metabolism of (R)- and (S)-methadone
in vitro. Drug Metab Dispos 2003;31:742-7.
19. Begre S, von Bardeleben U, Ladewig D, Jaquet-Rochat
S, Cosendai-Savary L, Golay KP, et al. Paroxetine in-
creases steady-state concentrations of (R)-methadone in
CYP2D6 extensive but not poor metabolizers. J Clin
Psychopharmacol 2002;22:211-5.
20. Israel BC, Blouin RA, McIntyre W, Shedlofsky SI. Ef-
fects of interferon-alpha monotherapy on hepatic drug
metabolism in cancer patients. Br J Clin Pharmacol 1993;
36:229-35.
21. Horsmans Y, Brenard R, Geubel AP. Short report:
interferon-alpha decreases 14C-aminopyrine breath test
values in patients with chronic hepatitis C. Aliment Phar-
macol Ther 1994;8:353-5.
22. Pageaux GP, le Bricquir Y, Berthou F, Bressot N, Picot
MC, Blanc F, et al. Effects of interferon-alpha on cyto-
chrome P-450 isoforms 1A2 and 3A activities in patients
with chronic hepatitis C. Eur J Gastroenterol Hepatol
1998;10:491-5.
23. Islam M, Frye RF, Richards TJ, Sbeitan I, Donnelly SS,
Glue P, et al. Differential effect of IFNalpha-2b on the
cytochrome P450 enzyme system: a potential basis of
IFN toxicity and its modulation by other drugs. Clin
Cancer Res 2002;8:2480-7.
24. Duclos-Vallee JC, Nishioka M, Hosomi N, Arima K,
Leclercq A, Bach JF, et al. Interferon therapy in LKM-1
positive patients with chronic hepatitis C: follow-up by a
quantitative radioligand assay for CYP2D6 antibody de-
tection. J Hepatol 1998;28:965-70.
25. Dalekos GN, Wedemeyer H, Obermayer-Straub P, Kay-
ser A, Barut A, Frank H, et al. Epitope mapping of
cytochrome P4502D6 autoantigen in patients with
chronic hepatitis C during alpha-interferon treatment.
J Hepatol 1999;30:366-75.
26. Becquemont L, Chazouilleres O, Serfaty L, Poirier JM,
Broly F, Jaillon P, et al. Effect of interferon alpha-
ribavirin bitherapy on cytochrome P450 1A2 and 2D6
and N-acetyltransferase-2 activities in patients with
CLINICAL PHARMACOLOGY & THERAPEUTICS
2005;77(3):214-24 Peginterferon alfa-2a and methadone PK 223
chronic active hepatitis C. Clin Pharmacol Ther
2002;71:488-95.
27. Geber WF, Lefkowitz SS, Hung CY. Effect of morphine,
hydromorphone, methadone, mescaline, trypan blue, vi-
tamin A, sodium salicylate, and caffeine on the serum
interferon level in response to viral infection. Arch Int
Pharmacodyn Ther 1975;214:322-7.
28. Quinn DI, Wodak A, Day RO. Pharmacokinetic and
pharmacodynamic principles of illicit drug use and treat-
ment of illicit drug users. Clin Pharmacokinet 1997;33:
344-400.
29. Modi MW, Fried M, Reindollar RW, Rustgi VR, Kenny
R, Wright TL, et al. The pharmacokinetic behavior of
pegylated (40KDA) interferon alfa-2a (Pegasys) in
chronic hepatitis C patients after multiple dosing [ab-
stract]. Hepatology 2000;32:394A.
30. Martin NE, Modi MW, Reddy R. Characterization of
pegylated (40KDA) interferon alfa-2a (Pegasys) in the
elderly [abstract]. Hepatology 2000;32:348A.
31. Xu Z-X, Hoffman J, Patel I, Joubert P. Single-dose
safety/tolerability and pharmacokinetic/pharmacodynam-
ics (PK/PD) following administration of ascending sub-
cutaneous doses of pegylated-interferon (PEG-IFN) and
interferon alfa-2a (IFN alfa-2a) to healthy subjects [ab-
stract]. Hepatology 1998;28:702A.
32. Perry CM, Jarvis B. Peginterferon-alpha-2a (40 kD): a
review of its use in the management of chronic hepatitis
C. Drugs 2001;61:2263-88.
33. Yan S. Clinical study reportProtocol NP16569: A
comparability study of peginterferon (40kDa) alfa-2a of
clinical trial material and the commercial product admin-
istered subcutaneously via needle injection in healthy
subjects. Nutley (NJ): Roche; 2002 Mar 29. Research
report 1007463.
34. Zeuzem S, Herrmann E, Lee JH, Fricke J, Neumann AU,
Modi M, et al. Viral kinetics in patients with chronic
hepatitis C treated with standard or peginterferon
alpha2a. Gastroenterology 2001;120:1438-47.
35. Zeuzem S, Feinman V, Rasenack J, Heathcote EJ, Lai
MY, Gane E, et al. Peginterferon alfa-2a in patients with
chronic hepatitis C. N Engl J Med 2000;343:1666-72.
36. Van Thiel DH, Anantharaju A, Creech S. Response to treat-
ment of hepatitis C in individuals with a recent history of
intravenous drug abuse. Am J Gastroenterol 2003;98:2281-8.
37. Sylvestre DL. Treating hepatitis C in methadone main-
tenance patients: an interim analysis. Drug Alcohol De-
pend 2002;67:117-23.
38. Backmund M, Meyer K, Von Zielonka M, Eichenlaub D.
Treatment of hepatitis C infection in injection drug users.
Hepatology 2001;34:188-93.
CLINICAL PHARMACOLOGY & THERAPEUTICS
224 Sulkowski et al MARCH 2005
LETTERS TO THE EDITOR
Disposition of cisapride appears to be inuenced
by P-glycoprotein in the mouse
To the Editor:
Recently, Lowry et al
1
evaluated cisapride as a potential
model substrate to assess cytochrome P450 (CYP) 3A4 ac-
tivity in vivo. As part of their evaluation, they conducted
in vitro studies to determine whether cisapride is a substrate
of P-glycoprotein (P-gp). These studies were conducted with
the use of LLC-PK1 cells and the derivative cell line
L-MDR1. Results from these studies suggest that cisapride is
not a substrate of P-gp. However, cisapride markedly in-
creased the accumulation of vinblastine in L-MDR1 cells,
with an apparent inhibition constant of approximately 16.1
mol/L. Cisapride (20 mol/L) also inhibited vinblastine
transport from the basal to the apical side by approximately
30% from baseline. These in vitro data suggest that cisapride
inhibits P-gp.
We evaluated the disposition of cisapride (Janssen Phar-
maceutica NV, Beerse, Belgium) in mdr1a/1b (/) mice
after oral administration of a 5-mg/kg dose and determined
the plasma and brain (whole-brain homogenates) concen-
trations of cisapride at 0.5, 1, 2, and 5 hours. Previous
studies have shown that the bioavailability and the brain
penetration of drugs that are substrates of P-gp are en-
hanced in mice lacking P-gp expression,
2
because P-gp is
highly expressed in the capillary endothelial cells of the
blood-brain barrier and the epithelial cells of the intes-
tine.
3,4
The area under the concentration versus time curve
from 0 to 5 hours [AUC(0-5)] of cisapride in the brains of
mdr1a/1b (/) was 5-fold higher in the mdr1a/1b (/)
mice compared with the mdr1a/1b (/) mice (316
ng h/mL versus 62.9 ng h/mL) (Fig 1). In contrast, the
AUC(0-5) of cisapride in the plasma of mdr1a/1b (/)
was 1.3-fold higher in the mdr1a/1b (/) mice compared
with the mdr1a/1b (/) mice (121 ng h/mL versus
90.2 ng h/mL) (Fig 1). With normalization for plasma
concentrations of cisapride, the brain-to-plasma area under
the concentration versus time curve (AUC) ratio of cisa-
pride in mdr1a/1b (/) mice was 2.6 compared with 0.7
in mdr1a/1b (/) mice; a similar observation was made
with maximum plasma concentration (C
max
) (Fig 2).
Our ndings suggest that, in vivo, cisapride appears to be
a moderate substrate of P-gp and that the brain penetration of
cisapride is enhanced in mice lacking P-gp expression. How-
ever, P-gp expression in the intestine appears to have minimal
inuence on the disposition of cisapride, possibly as a result
of both the inhibition and the saturation of P-gp at the level of
the intestine by cisapride. Although cisapride is no longer
marketed in the United States because of cardiac safety con-
cerns (arrhythmias),
5
it continues to be used as an orally
administered prokinetic agent for the treatment of gastro-
esophageal reux disease in many European and Asian coun-
tries. These ndings suggest not only that CYP3A4-based
interactions should be considered when drugs are coadminis-
tered with cisapride but also that caution should be exercised
when drugs that may be substrates or inhibitors of P-gp are
coadministered.
Edna F. Choo, PhD
Karthik Venkatakrishnan, PhD
Fig 1. Plasma (squares) and brain (triangles) levels of cisa-
pride in mdr1a/1b (/) (solid symbols) and mdr1a/1b
(/) (open symbols) mice. Data represent mean (SD)
values of 4 or 5 determinations at each time point.
Fig 2. Mean brain-to-plasma area under the concentration
versus time curve (AUC) and maximum plasma concentration
(C
max
) (SD) ratios in mdr1a/1b (/) (solid bars) and
mdr1a/1b (/) (open bars) mice.
CLINICAL PHARMACOLOGY & THERAPEUTICS MARCH 2005 225
Heather L. Hatch, MA
Sandhya Rahematpura, BSc
Pharmacokinetics, Dynamics and Metabolism
Pzer Inc
Groton, Conn
No conict of interest has been identied.
References
1. Lowry JA, Kearns GL, Abdel-Rahman SM, Nafziger AN, Khan
IS, Kashuba ADM, et al. Cisapride: a potential model substrate to
assess cytochrome P4503A4 activity in vivo. Clin Pharmacol Ther
2003;73:209-22.
2. Kim RB, Fromm MF, Wandel C, Leake B, Wood AJJ, Roden DM,
et al. The drug transporter P-glycoprotein limits oral absorption
and brain entry of HIV-1 protease inhibitors. J Clin Invest 1998;
101:289-94.
3. Gottesman MM, Pastan I. Biochemistry of multidrug resistance
mediated by the multidrug transporter. Annu Rev Biochem 1993;
62:385-427.
4. Cordon-Cardo C, OBrien JP, Casals D, Rittman-Grauer L, Bied-
ler JL, Melamed MR, et al. Multidrug-resistance gene (P-
glycoprotein) is expressed by endothelial cells at blood-brain
barrier sites. Proc Natl Acad Sci U S A 1989;86:695-8.
5. Michalets EL, Williams CR. Drug interactions with cisapride:
clinical implications. Clin Pharmacokinet 2000;39:49-75.
doi:10.1016/j.clpt.2004.10.009
Modulation of celecoxib pharmacokinetics by food
in pediatric patients
To the Editor:
In the November 2002 issue of the Journal, we presented
preliminary pharmacokinetic results of a clinical trial evalu-
ating the use of celecoxib in conjunction with metronomic
dosing of chemotherapy for antiangiogenic treatment in chil-
dren with recurrent solid tumors.
1
The purpose of this report
is 2-fold: (1) to present new pharmacokinetic data for cele-
coxib in a larger group of children taking celecoxib at a dose
of 250 mg/m
2
twice daily and (2) to present the effect of
high-fat food on celecoxib kinetics in a subset of children.
Our data suggest a trend toward increased absorption and
increased overall systemic exposure when celecoxib is taken
with a high-fat meal or snack. In addition, it appears that food
high in fat tends to modulate trough plasma levels such that
patients spend a smaller proportion of the dosing interval
below the targeted antiangiogenic concentration of 500 g/L
(Dr Jaime L. Masferrer, oral communication, 2002).
Data presented in our original report and data that have been
accrued since that publication indicate that the trough concen-
trations of celecoxib in our patients were below the desired
concentration that possesses antiangiogenic properties (500
g/L) (Dr Jaime L. Masferrer, oral communication, 2002). For
safety and regulatory reasons, we opted to try to increase the
systemic exposure to celecoxib by coadministering the drug with
a high-fat meal or snack rather than increasing the dose. Early
clinical studies investigating the pharmacokinetics of celecoxib
in healthy adults showed that the maximum plasma concentra-
tion (C
max
) and area under the plasma concentrationtime curve
(AUC) were approximately dose-proportional across the dose
range of 100 to 200 mg. However, at higher doses (400 mg,
which is approximately equivalent to our study dose of 250
mg/m
2
), there appears to be a less than proportional increase in
C
max
and AUC. This is thought to be a result of the low
solubility of celecoxib in aqueous media.
2
Therefore it was
recommended that higher doses of celecoxib be coadministered
with food to try to improve absorption.
3
Our study was amended
Table I. Single-dose and steady-state pharmacokinetics of celecoxib administered with and without food in
pediatric cancer patients
No.
Dose
(mg)
t
max
(h)
C
max
(g/L)
C
min
(g/L)
AUC*
(g/L h)
V
d
/F
(L/kg)
Cl/F
(L h
1
kg
1
) t
(h)
Predicted
time of
500 g/
L (h)
Time
spent at
500 g/
L (h)
Single dose
Without food 20 350
(median)
3
(median)
1257 869 213 114 7,810 3,484 8.2 6.7 1.4 0.9 4.0 1.5
With food 7 300
(median)
3
(median)
2284 998 297 155 12,498 3,871 4.0 1.5 0.8 0.2 3.6 1.4
% Change 82% 39% 60% 51% 43% 10%
P value .016 .138 .006 .115 .033 .599
Steady state
Without food 16 300
(median)
3
(median)
1440 756 296 162 8,886 4,521 6.8 4.9 0.9 0.4 5.0 2.4 8.1 3.6 4.1 3.2
With food 7 300
(median)
3
(median)
2867 1581 394 271 15,574 6,774 2.9 2.1 0.7 0.2 3.4 0.9 10.5 3.0 2.2 1.4
% Change 99% 33% 75% 57% 22% 32% 30% 46%
P value .007 .289 .011 .063 .103 .105 .147 .154
Data are expressed as mean SD. Comparisons within the single-dose and steady-state data were made by use of the Student t test with the signicance level set
at P .05.
t
max
, Time to maximum plasma concentration; C
max
, maximum plasma concentration; C
min
, trough concentration; AUC, area under plasma concentrationtime curve;
t