Cytochrome P450 2D6 (CYP2D6) comprises 2% to6% of the total hepatic cytochrome P450 (CYP) content1

but is responsible for the metabolism of many importantmedications.2 Dextromethorphan has been used exten-sively for CYP2D6 phenotyping since it was shown thatthe metabolism of dextromethorphan cosegregated withthat of the CYP2D6 probe debrisoquin (INN, debriso-quine).3 The main reasons for the use of dextromethor-

phan as a CYP2D6 probe are safety and availability. Todate, dextromethorphan phenotyping in adults has pri-marily used a 30-mg dose. However, data that supportthe use of any particular dextromethorphan dose forCYP2D6 phenotyping are lacking. One study that com-pared different dextromethorphan doses in one extensivemetabolizer concluded that no difference existed.4

Peak plasma concentrations of dextromethorphanafter oral doses of 30 to 120 mg to CYP2D6 extensivemetabolizers are <0.1 µmol/L,5-9 more than tenfold lessthan the Michaelis-Menten constant (Km) reported fordextromethorphan O-demethylation by CYP2D6 (1 to13 µmol/L).10,11 Even with large doses (2.5 mg/kg),plasma dextromethorphan concentrations were notdetectable (>0.21 µmol/L) in most subjects.12 There issome indication that dextromethorphan concentrationsin the liver during the first pass may be substantiallyhigher than peak plasma concentrations. Specifically,dextromethorphan undergoes considerable first-passhepatic metabolism, primarily through CYP2D6-mediated O-demethylation.6,7,9Similar characteristicsof imipramine led to the discovery of saturation in itsCYP2D6-mediated first-pass hepatic metabolism.13

The in vitro study by von Moltke et al10 showed thatalthough CYP2D6 was responsible for >70% of dex-

Dose dependency of dextromethorphanfor cytochrome P450 2D6 (CYP2D6)phenotyping

Most dextromethorphan CYP2D6 phenotyping studies use a 30-mg dose, but data that show superiorityof any particular dose are lacking. We compared metabolic ratios from six different dextromethorphanphenotyping doses to ascertain whether linearity existed over a dosage range. Forty subjects were enrolledin the study. Each subject received 0.05 mg/kg, 0.15 mg/kg, 0.3 mg/kg, 30 mg, 0.8 mg/kg, and 1.2mg/kg dextromethorphan in a randomized crossover fashion. Urinary dextromethorphan to dextrorphanmolar ratios were used to measure CYP2D6 activity. Single blood samples were obtained for CYP2D6genotyping, which revealed one poor metabolizer and 39 extensive metabolizers. A statistical differencewas found for the molar ratio between the 0.8 mg/kg and the 1.2 mg/kg dose compared with the otherfour doses. None of the 39 genotypic extensive metabolizers were incorrectly phenotyped with any of thesedoses. These data support the use of moderate doses of dextromethorphan for phenotyping to avoid dosedependency. (Clin Pharmacol Ther 1999;66:535-41.)

Daniel S. Streetman, PharmD, Ross E. Ellis, MD, Anne N. Nafziger, MD, MHS,J. Steven Leeder, PharmD, PhD, Andrea Gaedigk, PhD, Russell Gotschall, MS,Gregory L. Kearns, PharmD, and Joseph S. Bertino, Jr, PharmDCooperstown, NY, and Kansas City, Mo

From the Clinical Pharmacology Research Center, the Departmentof Medicine, and the Department of Pharmacy Services, BassettHealthcare, Cooperstown, and the Division of Pediatric ClinicalPharmacology and Experimental Therapeutics, Children’s MercyHospital and Clinics, and the Departments of Pediatrics and Phar-macology, University of Missouri–Kansas City.

Presented in part at the American Society for Clinical Pharmacologyand Therapeutics, San Antonio, Texas, 1999.

Supported in part by the E. Donnall Thomas Resident ResearchProgram in Internal Medicine and by grant 2 U01 HD31313-07,Network of Pediatric Pharmacology Research Units, NationalInstitute of Child Health and Human Development, Bethesda, Md.

Received for publication May 24, 1999; accepted Aug 17, 1999.Reprint requests: Joseph S. Bertino, Jr, PharmD, Clinical Pharma-

cology Research Center, Bassett Healthcare, One Atwell Road,Cooperstown, NY 13326-1394. E-mail: jbertino@iex.net

Copyright © 1999 by Mosby, Inc.0009-9236/99/$8.00 + 013/1/102333

535

tromethorphan O-demethylation at dextromethorphanconcentrations <5 µmol/L, as dextromethorphan con-centrations increase, the importance of CYP2D6declines and CYP2C9 becomes primarily responsiblefor dextrorphan formation. If these observations holdtrue in vivo, lower dextromethorphan doses may pro-vide a more accurate reflection of actual CYP2D6activity. The primary goal of this study was to deter-mine whether the urinary dextromethorphan metabolicratio is dose dependent over a 24-fold dynamic range.

METHODSThis study was approved by the Institutional Review

Board of Bassett Healthcare, and written informed con-sent was obtained from all subjects.

Subjects. Subjects were recruited by advertisement andfrom a list of subjects enrolled in previous phenotypingtrials. Before beginning the study, all subjects underwenta screening medical history, including a complete reviewof their medical records and medication histories. All sub-jects were healthy adults (≥18 years old) who were notreceiving any known inhibitors or substrates of CYP2D6.Each subject was required to limit his or her alcohol con-sumption to one beer or less (or equivalent) per day forthe duration of the study. Women who were not surgicallysterile were required to use an acceptable barrier methodof birth control and underwent qualitative urine preg-nancy testing (Clear Blue Easy; Whitehall Laboratories,Madison, NJ) before each study phase. Women using oralcontraceptives were excluded from the study.

Phenotyping procedure. Before each oral dose ofdextromethorphan (Robitussin Pediatric Cough; AHRobins, Madison, NJ), subjects were instructed to com-pletely empty their bladders. Dextromethorphan dosesof 0.05 mg/kg, 0.15 mg/kg, 0.3 mg/kg, 30 mg, 0.8mg/kg, and 1.2 mg/kg were then administered in a pre-determined randomly assigned order at 8 PM on studydays. During each study period, subjects were instructedto collect all of their urine for the following 12 hours.Duration of urine collection and total urine volume wererecorded for each sample. Aliquots (approximately 15mL each) were collected from each 12-hour urine sam-ple and stored at –80°C until analysis. Each subjectreceived each of the six dextromethorphan doses, witha 2-week washout period between doses.

Urine assay for dextromethorphan and metabolites.Urine aliquots were assayed for concentrations of dex-tromethorphan and dextrorphan using a modification*

of the methods of Park et al13 and Lam and Rodriguez.14

In brief, 100 standard mU of β-glucuronidase-arylsul-fatase was added to each 3-mL aliquot of urine, whichwas subsequently placed into a shaking incubator for18 hours at 37°C. Internal standard (levallorphan tar-trate) was then added, and the pH was adjusted to 12before solid-phase extraction with Chem Elut columns(Varian Sample Preparation Products, Harbor City,Calif). Eluants were back extracted with 0.01 mol/Lhydrochloric acid, and the aqueous layer was dried ina Savant SpeedVac (Holbrook, New York, NY) at 50°C.Samples were then redissolved in 100 µL of 0.01Nhydrochloric acid, and 75 µL was injected into theHPLC system (HP model 1100 chromatographic sys-tem with a HP model 1046A fluorescence detector,Hewlett-Packard Co, San Fernando, Calif). Chromatog-raphy was performed at 25°C on a Novapak phenyl col-umn (Waters Corp, Milford, Mass) with use of a mobilephase that consisted of 20 mmol/L potassium phos-phate/hexane sulfonic acid (60%, pH 4.0), and acetoni-trile (40%) pumped at 1.2 mL/min with monitoring atexcitation and emission wavelengths of 235 and 310nm, respectively. Data output was normalized to theinternal standard, and the molar amount was deter-mined with use of standard curves prepared daily indrug-free urine specimens.

CYP2D6 genotype analysis. CYP2D6 genotypingwas performed on all subjects. A single 7-mL bloodsample was collected from the antecubital vein forCYP2D6genotyping. Blood samples were collectedinto Vacutainer tubes (Becton Dickson and Company,Franklin Lakes, NJ) with 1.5 mL acid-citrate-dextrose(ACD) solution A (trisodium citrate, 22.0 g/L; citricacid, 8.0 g/L; and dextrose, 24.5 g/L) and stored at 4°Cuntil analysis.

Genotyping of CYP2D6 was performed accordingto the methods of Gaedigk et al.† In brief, highly puregenomic deoxyribonucleic acid suitable for extra-longpolymerase chain reaction (XL-PCR) was preparedfrom peripheral blood mononuclear cells with theQIAmp blood kit (Qiagen, Chatsworth, Calif). To dis-tinguish between the CYP2D6gene and the CYP2D8pseudogene and the nonfunctional CYP2D7gene (bothlocated upstream from the CYP2D6gene), a primerpair was used to amplify the entire coding region ofCYP2D6by XL-PCR generating a 5.1 kb product. TheCYP2D6-specific XL-PCR product subsequentlyserved as a template for a series of polymerase chain

536 Streetman et alCLINICAL PHARMACOLOGY & THERAPEUTICS

NOVEMBER 1999

*Gotschall RR, Gaedigk A, Simon SD, Kearns GL, Leeder JS.Intraindividual variability in CYP2D6 activity: stability of the uri-nary dextromethorphan metabolic ratio and detection of drug-druginteractions. Pharmacogenetics 1999 [submitted for publication].

†Gaedigk A, Gotschall RR, Forbes NS, Simon SD, Kearns GL,Leeder JS. Misclassification analysis using an algorithm forcytochrome P4502D6 (CYP2D6) phenotype assignment from geno-type data. Pharmacogenetics 1999 [submitted for publication].

reaction–restriction fragment length polymorphismreactions designed to detect nucleotide point muta-tions, deletions, or insertions compared with the func-tional CYP2D6*1allele. The digestion products wereanalyzed by gelTWIN electrophoresis, stained withethidium bromide and visualized with a FluorImagerscanning fluorimeter (Molecular Dynamics, Sunny-vale, Calif). All subjects were tested for theCYP2D6*2, *3, *4, *5, *6, *7, *9, *10, *13/16, *17,and duplicate alleles according to an algorithm basedon published allele frequencies.†

Classification of metabolizer status. Each subjectwas classified as an extensive metabolizer or a poormetabolizer according to both CYP2D6 phenotype andgenotype. The dextromethorphan/dextrorphan molarratio after administration of an oral dose of dex-tromethorphan was used to phenotype subjects as exten-sive metabolizers or poor metabolizers.3 An antimodeof 0.3 was used to separate the extensive and poor phe-notypes. By genotype, extensive metabolizers weredefined by the presence of one or more functional (*1,*2, *10, or *17) CYP2D6alleles, whereas poor metab-olizers lacked two or more functional CYP2D6alleles.

Statistical analysis. All analyses were performed withthe SAS software system version 6.08 (SAS Institute,Cary, NC) on an ALPHA VMS host. The limit of signif-icance accepted for all statistical analyses was P ≤ .05.

Sample size calculations were performed a prioribased on data from Kashuba et al15 that quantifiedintraindividual variability in CYP2D6 activity over a3-month period. Because dextromethorphan/dextror-phan molar ratios were found to vary by a mean of62.7% within individuals, we determined that anypotential clinically significant change in activity wouldbe greater than “normal” intraindividual variability. Forthis study, a significant change in CYP2D6 activity wasdefined as a ≥100% change in the dextromethorphan/dextrorphan molar ratio. With use of an α of .05 and80% power (β = .2), approximately 37 subjects wouldbe required to detect a 100% change in the dex-tromethorphan/dextrorphan molar ratio.

Characteristics of the distribution of dextromethor-phan/dextrorphan molar ratios were investigated bymeasures of skewness and kurtosis. Because the datawere not normally distributed, molar ratios were log-transformed to obtain a normal distribution. Differencesin the transformed dextromethorphan/dextrorphanmolar ratio among the six different dextromethorphan

doses were determined with a generalized linear model,repeated-measures ANOVA. This method of analysistakes into account the various doses and the intrasub-ject variability for the six dosing levels. If a significantdifference was found, the Scheffe test was applied todetermine when the statistical difference occurred. Dataare presented as mean values ± SD.

RESULTSForty-two white subjects were enrolled into the study,

with 40 subjects completing all six study phases. Onesubject withdrew from the study for scheduling reasonsbefore receiving any of the six dextromethorphan doses.A second subject (CYP2D6*4/*4genotype) withdrewas a result of significant adverse effects (dizziness,drowsiness, and disorientation) after receiving one dex-tromethorphan dose (1.2 mg/kg). Only data obtainedfrom the 40 subjects who completed all six study phaseswere included in the analyses. Table I summarizes thedemographic information of the subjects.

Genotyping results revealed 26 subjects with twofunctional alleles, two subjects with more than twofunctional alleles, 11 subjects with only one functionalallele, and one subject without any functional alleles.All (39 of 39) genotypic extensive metabolizers werecorrectly identified by phenotyping with each of six dif-ferent dextromethorphan doses. The one genotypic poormetabolizer was correctly identified by phenotypingwith each dextromethorphan dose except for 0.15mg/kg dextromethorphan, which yielded a molar ratioof 0.166, classifying the subject as an extensive metab-olizer on that occasion. Allelic frequencies for the 40subjects are listed in Table II. Because only one poormetabolizer was identified in our population, the dex-

Streetman et al 537CLINICAL PHARMACOLOGY & THERAPEUTICSVOLUME 66, NUMBER 5

Table I. Demographic information on the 40 studysubjects

Demographic variable Value*

Age (y) 39 ± 6.5Sex

Men 10Women 30

Smoking statusNonsmokers 30Smokers 10

Height (cm) 168 ± 7.9Weight (kg) 75 ± 13.9Ideal body weight (kg) 60 ± 8.5Total body weight above ideal weight (%) 125.6 ± 21.5

*Values are listed as mean values ± SD when appropriate.

†Gaedigk A, Gotschall RR, Forbes NS, Simon SD, Kearns GL,Leeder JS. Misclassification analysis using an algorithm forcytochrome P4502D6 (CYP2D6) phenotype assignment from geno-type data. Pharmacogenetics 1999 [submitted for publication].

tromethorphan molar ratio statistical analysis was con-fined to the 39 extensive metabolizers.

The mean dextromethorphan molar ratios obtainedafter administration of each of the six dextromethor-phan doses according to genotype are summarized inTable III. Comparison of mean log-transformed dex-tromethorphan molar ratios with a generalized linearmodel and the Scheffe test revealed statistically signif-icant differences between the 0.8 and the 1.2 mg/kgdextromethorphan doses versus the 0.05 mg/kg, 0.15mg/kg, 0.3 mg/kg, and 30 mg doses (P ≤ .05). This sug-gests that the two larger doses show dose-dependentpharmacokinetics. Considerably more intersubject vari-ability in molar ratios among CYP2D6 extensivemetabolizers was observed after the largest versus the

smallest dextromethorphan dose (620-fold versus 45-to 88-fold). Fig 1 illustrates the mean molar ratio byCYP2D6genotype after each dextromethorphan dose.

Considerable differences in intrasubject variabilityduring the 3-month study period were also evident. Dex-tromethorphan molar ratios varied by a mean magnitudeof 6.2-fold, with a mean coefficient of variation (CV%)of 50.8%, within individual subjects during the studyperiod. Both intrasubject magnitude of variability andCV% varied significantly between individuals. Magni-tude of variability ranged from 1.6- to 41-fold, and CV%ranged from 17.2% to 121.6%. The dextromethor-phan/dextrorphan ratio varied less than 4-fold in themajority (55%) of subjects. In contrast, a greater than8-fold variability was observed in nine (22.5%) subjects.

Six of the 41 subjects who received at least one doseof dextromethorphan (14%) reported adverse effects. Onesubject dropped out of the study because of significantdizziness, drowsiness, and disorientation after adminis-tration of the highest (1.20 mg/kg) dextromethorphandose. Of the remaining subjects who reported adverseeffects, four of the five reported the occurrence of adverseeffects with only the highest dextromethorphan dose, andnone of those subjects required medical attention. Theremaining subject (*1/*2) reported altered dreams afterall six dextromethorphan doses. Table IV lists the reportedadverse effects and their incidences in this study.

538 Streetman et alCLINICAL PHARMACOLOGY & THERAPEUTICS

NOVEMBER 1999

Table II. CYP2D6allele frequency in 40 healthyvolunteers

CYP2D6 allele Frequency

*1 0.413*2 0.375*4 0.138*5 0.025*10 0.025*2x2 0.025

Fig 1. Illustration of mean dextromethorphan/dextrorphan (DM/DX) molar ratio for sixdextromethorphan doses according to CYP2D6genotype. Trianglesrepresent dextromethorphan/dextrorphan ratio for the CYP2D6 poor metabolizer (*4/*5); squaresrepresent mean dextromethor-phan/dextrorphan ratio for 39 extensive metabolizer. Broken line represents antimode(dextromethorphan/dextrorphan ratio = 0.30) that separates CYP2D6 extensive metabolizers andpoor metabolizers.

DISCUSSIONMost dextromethorphan phenotyping procedures use

a 30-mg dose, despite a lack of data to support this doseand the possibility of saturable pharmacokinetic para-meters.16-21 In our study, a significant difference wasobserved in the mean dextromethorphan molar ratiobetween the 0.8 mg/kg and 1.2 mg/kg dose versus theother four doses used in 39 CYP2D6 extensive metab-olizers. These data are pertinent to both the in vitro andin vivo results of studies of other CYP2D6 substratesthat suggest dose-dependent metabolism.10,17-21

Data obtained with the CYP2D6 phenotyping probedebrisoquin17 revealed evidence of dose dependency inCYP2D6 poor metabolizers. With use of the establishedantimode of 12.6 to distinguish poor metabolizers fromextensive metabolizers, the ability to discriminateextensive metabolizers from poor metabolizersappeared to be reduced with lower doses. The study bySloan et al17 was conducted in poor metabolizers, andtherefore the reported results were not related toCYP2D6 activity.

Studies of other known CYP2D6 substrates have indi-cated the possibility of nonlinear CYP2D6-mediateddrug metabolism. A study of propafenone-produced β-blockade by Lee et al19 suggested saturable metabolismof CYP2D6. There was evidence of significantly moreβ-blockade in poor metabolizers than in extensivemetabolizers at both the 150-mg and 225-mg doses, withno significant differences found at the 300-mg dose,possibly because of saturable metabolism in extensivemetabolizers at that dose. Disproportional increases inplasma propafenone concentrations with increasing dosesupport the theory of a saturable metabolic pathway.19-21

However, in vitro data suggest that CYP2D6 is respon-sible for only about 50% of propafenone metabolism,22

making it difficult to draw conclusions about CYP2D6saturation from studies of propafenone pharmacody-namics and pharmacokinetics.

Saturation of CYP2D6 during the first pass ofimipramine has also been proposed.16,18After adminis-tration of 50 mg intravenous imipramine and 100 mg oralimipramine to eight extensive metabolizers and threepoor metabolizers, the role of CYP2D6 in imipraminemetabolism decreased from 60% of total imipraminemetabolism during intravenous administration to 25%during oral administration,18 suggesting saturation ofCYP2D6 at the imipramine concentrations present dur-ing the first pass.16,18

A recent study by von Moltke et al10 has identifiedseveral CYP isoforms involved in the metabolic con-version of dextromethorphan to dextrorphan in vitro. Itwas estimated that CYP2D6 was primarily responsiblefor dextrorphan formation, accounting for 83.1% ofdextrorphan formation. CYP2C9 and CYP2C19 wereestimated to account for 13.3% and 3.6% of dextror-phan formation, respectively. With dextromethorphanconcentrations of <5 µmol/L, CYP2D6 is predicted toaccount for >70% of dextrorphan formation. As dex-tromethorphan concentrations increase, however,CYP2C9 becomes the predominant CYP isoformresponsible for dextrorphan formation, and its impor-tance exceeds that of CYP2D6 as concentrations exceed30 to 40 µmol/L.

Although peak plasma concentrations of dex-tromethorphan after oral doses of 30 tp 120 mg areconsiderably less than the reported Km of 1 to 13µmol/L for dextromethorphan O-demethylation byCYP2D6,9-15 it is unclear whether hepatic concentra-tions during first pass are sufficiently high that otherCYP isoforms significantly contribute to dextrorphan

Streetman et al 539CLINICAL PHARMACOLOGY & THERAPEUTICSVOLUME 66, NUMBER 5

Table III. Mean ± SD dextromethorphan molar ratio(dextromethorphan/dextrorphan) in 40 healthy volun-teers according to dextromethorphan dose andCYP2D6genotype*

Extensive metabolizers Poor metabolizer(n = 39) (*4/*5)

0.05 mg/kg 0.005 ± 0.005 2.4990.15 mg/kg 0.009 ± 0.017 0.1660.30 mg/kg 0.011 ± 0.018 1.49430 mg 0.009 ± 0.013 0.6450.80 mg/kg 0.012 ± 0.017 6.8061.20 mg/kg 0.015 ± 0.027 0.619

*Molar ratios reported in this table are not log-transformed.

Table IV. Reported adverse effects during the study

Adverse effect* Number (%)

Dizziness 4 (9.8)Nausea 3 (7.3)Drowsiness 2 (4.9)Impaired concentration 1 (2.4)Disorientation 1 (2.4)Lightheadedness 1 (2.4)Anterograde memory loss 1 (2.4)Vomiting 1 (2.4)Diarrhea 1 (2.4)Altered dreams† 1 (2.4)

*All reported adverse effects except altered dreams were reported with onlythe largest dextromethorphan dose (1.2 mg/kg).

†Altered dreams were reported for all six dextromethorphan doses in onesubject.

formation. Saturation of CYP2D6 during hepatic first-pass has been reported for imipramine despite plasmaconcentrations that are substantially lower thanreported Km values for CYP2D6-mediated imipraminemetabolism.5,17

We found that relatively larger doses (0.8 mg/kg or56 mg in a 70-kg adult and 1.2 mg/kg or 84 mg in a 70-kg adult) gave a statistical difference in dextromethor-phan molar ratio found versus doses of 0.05 mg/kg,0.15 mg/kg, 0.3 mg/kg, and 30 mg. This suggests dosedependency. Despite this difference, all 39 genotypicextensive metabolizers were phenotyped as extensivemetabolizers. The one poor metabolizer (*4/*5) sub-ject was phenotyped as a poor metabolizer by five ofthe six dextromethorphan doses (molar ratios, 0.619 to6.806). Phenotyping with 0.15 mg/kg dextromethor-phan resulted in an extensive metabolizer molar ratio(0.166) in this subject. This discrepancy is probably theresult of normal intrasubject variability in the dex-tromethorphan molar ratio, which has been shown tobe considerable in some cases. We believe that this dis-cordant result is attributable to normal variability andnot to an incorrect genotype assignment or to errors thedextromethorphan/dextrorphan urine assays becausethe other five phenotypes were consistent with the poormetabolizer genotype.

In conclusion, this study has showed statistically sig-nificant differences in dextromethorphan molar ratiosobtained after dextromethorphan doses of 0.8 mg/kgand 1.2 mg/kg. As a result of these findings, investiga-tors could use a dose of dextromethorphan from 0.05mg/kg to 0.3 mg/kg or the commonly used dose of 30mg for CYP2D6 phenotyping without concern aboutthe ability of metabolic ratio to determine phenotype.These findings may be particularly relevant to pheno-typing studies of pediatric populations, when use of therecommended adult dose may not be desirable or whensubjects may not ingest the entire administered dose.

References

1. Shimada T, Yamazaki H, Mimura M, Inui Y, GuengerichFP. Interindividual variations in human liver cytochromeP-450 enzymes involved in the oxidation of drugs, car-cinogens and toxic chemicals: studies with liver micro-somes of 30 Japanese and 30 Caucasians. J PharmacolExp Ther 1994;270:414-23.

2. Rendic S, Di Carlo FJ. Human cytochrome P450enzymes: a status report summarizing their reactions, sub-strates, inducers, and inhibitors. Drug Metab Rev1997;29:413-580.

3. Schmid B, Bircher J, Preisig R, Kupfer A. Polymorphicdextromethorphan metabolism: co-segregation of oxida-

tive O-demethylation with debrisoquin hydroxylation.Clin Pharmacol Ther 1985;38:618-24.

4. Kupfer A, Schmid B, Pfaff G. Pharmacogenetics of dex-tromethorphan O-demethylation in man. Xenobiotica1986;16:421-33.

5. Schadel M, Wu H, Otton SV, Kalow W, Sellers EM. Phar-macokinetics of dextromethorphan and metabolites inhumans: influence of the CYP2D6 phenotype and quini-dine inhibition. J Clin Psychopharmacol 1995;15:263-9.

6. Vetticaden SJ, Cabana BE, Prasad VK, Purich ED, JonkmanJHJ, de Zeeuw R, et al. Phenotypic differences in dex-tromethorphan metabolism. Pharm Res 1989;6:13-9.

7. Capon DA, Bochner F, Kerry N, Mikus G, Danz C,Somogyi AA. The influence of CYP2D6 polymorphismand quinidine on the disposition and antitussive effect ofdextromethorphan in humans. Clin Pharmacol Ther 1996;60:295-307.

8. Mortimer O, Lindstrom B, Laurell H, Bergman U, RaneA. Dextromethorphan: polymorphic serum pattern of theO-demethylated and di-demethylated metabolites in man.Br J Clin Pharmacol 1989;29:223-7.

9. Zhang Y, Britto MR, Valderhaug KL, Wedlund PJ, SmithRA. Dextromethorphan: enhancing its systemic availabil-ity by way of low-dose quinidine-mediated inhibition ofcytochrome P4502D6. Clin Pharmacol Ther 1992;51:647-55.

10. von Moltke LL, Greenblatt DJ, Grassi JM, Granda BW,Venkatakrishnan K, Schmider J, et al. Multiple humancytochromes contribute to biotransformation of dex-tromethorphan in vitro: role of CYP2C9, CYP2C19,CYP2D6, and CYP3A. J Pharm Pharmacol 1998;50:997-1004.

11. Schmider J, Greenblatt DJ, Fogelman SM, von MoltkeLL, Shader RI. Metabolism of dextromethorphan in vitro:involvement of cytochromes P450 2D6 and 3A3/4, witha possible role of 2E1. Biopharm Drug Dispos 1997;18:227-40.

12. Hollander D, Pradas J, Kaplan R, McLeod HL, EvansWE, Smith RA. High-dose dextromethorphan in amyo-trophic lateral sclerosis: phase I safety and pharmacoki-netic studies. Ann Neurol 1994;36:920-4.

13. Park YH, Kullberg MP, Hinsvark ON. Quantitative deter-mination of dextromethorphan and three metabolites inurine by reverse-phase high-performance liquid chro-matography. J Pharm Sci 1984;73:24-9.

14. Lam YW, Rodriguez SY. High-performance liquid chro-matography determination of dextromethorphan and dex-trorphan for oxidation phenotyping by fluorescence andultraviolet detection. Ther Drug Monit 1993;15:300-4.

15. Kashuba ADM, Nafziger AN, Kearns GL, Leeder JS,Shirey CS, Gotschall R, et al. Quantification of three-month intraindividual variability and the influence ofmenstrual cycle phase on CYP2D6 activity as measuredby dextromethorphan phenotyping. Pharmacogenetics1998;8:403-10.

16. Brosen K. Recent developments in hepatic drug oxida-

540 Streetman et alCLINICAL PHARMACOLOGY & THERAPEUTICS

NOVEMBER 1999

tion: implications for clinical pharmacokinetics. ClinPharmacokinet 1990;18:220-39.

17. Sloan TP, Lancaster R, Shah RR, Idle JR, Smith RL.Genetically determined oxidation capacity and the dispo-sition of debrisoquine. Br J Clin Pharmacol 1983;15:443-50.

18. Brosen K, Gram LF. First-pass metabolism of imipramineand desipramine: impact of the sparteine oxidation phe-notype. Clin Pharmacol Ther 1988;43:400-6.

19. Lee JT, Kroemer HK, Silberstein DJ, Funck-Brentano C,Lineberry MD, Wood AJ, et al. The role of geneticallydetermined polymorphic drug metabolism in the beta-blockade produced by propafenone. N Engl J Med 1990;322:1764-8.

20. Siddoway LA, Thompson KA, McAllister CB, Wang T,Wilkinson GR, Roden DM, et al. Polymorphism ofpropafenone metabolism and disposition in man: clinicaland pharmacokinetic consequences. Circulation 1987;75:785-91.

21. Connolly SJ, Kates RE, Lebsack CS, Harrison DC, Win-kle RA. Clinical pharmacology of propafenone. Circula-tion 1983;68:589-96.

22. Botsch S, Gautier JC, Beaune P, Eichelbaum M, Kroe-mer HK. Identification and characterization of thecytochrome P450 enzymes involved in N-dealkylation ofpropafenone: molecular base for interaction potential andvariable disposition of active metabolites. Mol Pharma-col 1993;43:120-6.

Streetman et al 541CLINICAL PHARMACOLOGY & THERAPEUTICSVOLUME 66, NUMBER 5

Receive tables of contents by E-mail

To receive the tables of contents by E-mail, sign up through our Web site at

http://www.mosby.com/cpt

Choose “E-mail Notification.”Simply type your E-mail address in the box and click the “Subscribe” button.

Alternatively, you may send an E-mail message to majordomo@mosby.com. Leave the subjectline blank and type the following as the body of your message:

subscribe cpt_toc

You will receive an E-mail to confirm that you have been added to the mailing list. Note thattable of contents E-mails will be sent out when a new issue is posted to the Web site.