Determination of dextromethorphan metabolic phenotype by salivary analysis with a reference to genotype in Chinese patients receiving renal hemodialysis*

Determination of dextromethorphan metabolic phenotype by salivary analysis with a reference to genotype in Chinese patients receiving renal hemodialysis

Background: The polymorphic metabolism of debrisoquin and sparteine by cytochrome P450IID6 (CYP2D6) is genetically determined. Determination of the CYP2D6 metabolic phenotype with conven- tional urine analytic methods is not feasible in anuric patients with renal failure. The possibility of using salivary analysis, with dextromethorphan as a probe drug, to determine the CYP2D6 metabolic phenotype in patients with renal failure was evaluated. Methods and results: One hundred four Chinese patients with renal failure were recruited. All 104 patients were receiving hemodialysis. Saliva was collected before and at 3 hours after each patient took a capsule of dextromethorphan hydrobromide (30 mg). Four patients were excluded because of insufficient samples of saliva. The distribution of logarithms of the metabolic ratios (log[MR]) in the 100 patients appeared to be normal. Administration of quinidine sulfate (200 mg twice daily) to nine of the patients significantly and markedly increased the dextromethorphan metabolic ratios. The metabolic ratios of nine patients pretreated with quinidine were higher than any of the 100 patients with renal failure who did not receive quinidine pretreatment. A metabolic ratio of 33 separated these two groups. Genomic deoxyribonucleic acid was extracted from whole blood in a subset of patients. Polymerase chain reaction @‘CR)-based methods were used to detect the CYP2D6 and B mutant genes. Mutant B alleles (which are common in white poor metabolizers) of CYP2D6 genes were not detected in any of the 47 subjects tested. A PCR- based test of cytosine (C,,,) to thym’ me (Tiss) polymorphism at 188 base pairs in exon 1 of CYP2D6 genes was performed in 61 patients. Subjects who were homozygous for C,,, had significantly (p = 0.0067) lower log[MR] values than those who were homozygous for T,,,. Conclusions: Determination of dextromethorphan metabolic ratios in saliva is feasible in patients with renal failure requiring hemodialysis. All subjects in this study appeared to be “extensive metabolizer” phenotype for CYP2D6, and no poor metabolizer was identified. From the results with quinidine pretreatment, a metabolic ratio of 33 is suggested to be a tentative antimode for identification of poor metabolizers in patients with renal failure. ( CLIN P HARMACOL THER 1996;59:411-7.)

Zone-Yuan Hou, MD, Ching-Pu Chen, MD, Wu-Chang Yang, MD, Ming-Derg Lai, PhD, Elaine T. Buchert, PhD, Hsiao-Min Chung, MD, Linda W. Pickle, PhD, and Raymond L. Woosley, MD, PhD Kaohsiung and Tainan, Taiwan, Republic of China, Washington, D.C., and Atlanta, Ga.

From the Department of Medicine, Veterans General Hospital, Kaohsiung; the Department of Biochemistry, College of Medi- tine, National Cheng Kung University, Tainan; the Department of Pharmacology, Georgetown University Medical Center, Wash- ington; and the National Center for Health Statistics. U.S. De-

Oxidative drug metabolism by hepatic microso- ma1 CYP2D6 enzyme is genetically determined’ and has a polymorphic distribution in most populations. More than 40 drugs2-5 are known to be primarily

partment of Health and Human Services, Centers for Disease metabolized by this pathway. Genetically deter- Control, Atlanta.

Supported by grant number NSC 80-0412-B075-78 from the National Science Council, Taipei, Taiwan, Republic of China.

Received for publication June 16, 1995; accepted Sept. 21, 1995.

Reprint requests: Zone-Yuan Hou, MD, Department of Medi- tine, God’s Help Hospital, 66 Lane 601 Section 2, Pei-Kang Road, Tai-Pao City, Chia-Yi County, Taiwan, R.O.C.

13/l/69519

411

412 Hou et al. CLINICAL PHARMACOLOGY &THERAPEUTICS

APRIL 1996

mined variability in metabolism is partially responsible for interindividual variability of response to these drugs. Dextromethorphan is a substrate for the polymorphic CYP2D6 enzyme and is used widely to determine the metabolic phenotype. The phenotyping involves ingestion of a single oral dose of dextromethorphan, followed by an 8- to lo-hour urine collection.6-8 Because urine collection may not be feasible or reliable in patients with renal failure, an alternative method is required. In a previous study in normal subjects,’ analysis of saliva samples collected at 3 hours after taking dextromethorphan was found to be a good alternative to urinary analysis for determination of dextromethorphan metabolic phenotype. The purpose of this study was to validate the use of salivary analysis of dextromethorphan in patients with renal failure requiring hemodialysis. Furthermore, quinidine was shown to be a potent inhibitor of CYP2D6 enzyme.‘” Concomitant administration of quinidine and CYP2D6 substrates markedly increased the metabolic ratios1i,i2 and may even convert extensive metabolizers to poor metabolizers.13,i4 Therefore the effect of quinidine on the dextromethorphan metabolic ratios was tested and CYP2D6 genotypic analysis was performed to examine the validity of salivary analysis.

MATERIAL AND METHODS Patients

One hundred four patients with end-stage renal disease were recruited into the study. All 104 patients were receiving regular hemodialysis and were Chinese inhabitants of Taiwan. There were 74 men and 30 women; mean ? SD age was 56.0 ? 13.5 years and mean body weight was 55.0 + 8.0 kg. They were allowed to continue their medications. for hy- pertension, diabetes mellitus, and renal failure dur- ing the phenotyping. However, medications that were known to interact with dextromethorphan or CYP2D6 substrates were discontinued for at least 1 week before determination of metabolic phenotype. Written informed consent was obtained from each patient. The study was approved by the local ethics committee at Kaohsiung, Taiwan.

Protocol The patients were phenotyped while not under-

going hemodialysis. Saliva collection was performed as reported previously,’ except that the dose of dextromethorphan hydrobromide was reduced to 30 mg. Determination of the dextromethorphan metabolic ratio was repeated in nine patients after they

were treated with 200 mg quinidine sulfate twice daily for 7 days.

Phenotype analysis The concentrations of dextromethorphan and its

major metabolite dextrorphan in saliva were measured as described previously, with modification.’ In the present study, we modified the method so that the column was connected sequentially to a fluorescence detector (Waters 470, Millipore Corporation, Milford, Mass.) and an ultraviolet detector (Waters 484). Dex- tromethorphan and dextrorphan were detected with use of the fluorescence detector at wavelengths of 200/310 nm (excitation/emission). Quinidine was detected with the ultraviolet detector at a wavelength of 202 nm. Encainide hydrochloride (Bristol-Myers Co., Wallingford, Conn.) was used as the internal standard for the samples from the nine patients pretreated with quinidine. To separate quinidine from encainide, the mobile phase was modified to a 45:55 ratio of aceto- nitrile to monobasic potassium phosphate and was titrated to a pH of 3.3. The ultraviolet wavelength was set at 280 nm to detect quinidine and encainide.

Genotype analysis Detection for mutant B alleles. Blood samples of 47

patients were analyzed for the presence of mutant CYP2D6 gene, B alleles. This part of the analysis was performed in Dr. Woosley’s laboratory in Georgetown University. The details of the analysis has been described by Buchert et a1.15 In brief, 10 ml blood samples were collected into tubes with hepa- rin and stored at -70” C until analysis. Genomic deoxyribonucleic acid (DNA) was extracted from whole blood as described by Walsh et a1.i6 One tenth volume (0.02 ml) of the isolated DNA was analyzed for the presence of CYP2D6-wild type and CYP2D6-B (mutant) alleles. Polymerase chain reaction (PCR) analysis of the B alleles was performed as previously described by Hanioka et a1.l’

Diagnostic test of C/T,,, polymorphism. A PCR- based test of cytosine to thymine polymorphism at 188 base pairs in exon 1 of CYP2D6 genes (C/T,,, polymorphism) was established in Dr. Lai’s laboratory and has been described previously.” Sixty-one patients had blood samples tested for C/T,,, polymorphism using this method.

Statistics. Data are expressed as mean values + SD. Group comparisons of normally distributed data were performed with the Student t test. A p value less than 0.05 was required for rejection of the null hypothesis.

CLINICAL PHARMACOLOGY &THERAPEUTICS VOLUME 59, NUMBER 4

RESULTS Among the 104 patients who were asked to collect

saliva, four failed to produce sufficient samples for determination of drug concentrations and were excluded from further analysis. Three of the 104 patients reported symptoms of headache, dizziness, or fatigue after taking dextromethorphan. Among nine patients pretreated with quinidine sulfate, six reported symptoms that included vomiting (two patients), dizziness (three patients), epigastric pain (one patient), palpitations (one patient), and drows- iness (one patient).

The saliva concentrations of dextromethorphan and dextrorphan were undetectable in two and three patients, respectively, but none had undetectable concentrations of both compounds. For calculation of the metabolic ratios, undetectable concentrations were assumed to be 1 rig/ml (the lower limit of determination). The mean values of dextromethorphan and dextrorphan concentrations in saliva were 25 It 31 rig/ml (range, 1 to 216 nglml) and 11 2 9 rig/ml (range, 1 to 50 r&ml), respectively.

Metabolic ratios The distribution of logarithms of the metabolic ra-

tios (log[MR]) without pretreatment with quinidine in 100 patients is shown in Fig. 1, upperpanel. There were no obvious outliers, and the distribution appeared to be normal. In the nine patients who received quinidine, the mean saliva concentration of quinidine was 0.90 ? 1.15 Fdrnl. The influence of quinidine pretreatment on the metabolic ratios is shown in Table I. The metabolic ratios increased from a mean of 4.15 L 3.19 to 169.3 ? 207.6 (p < 0.05). The log[MR] values after quinidine pretreatment are shown in Fig. 1, lower panel, and were higher than those of patients who did not receive quinidine. A metabolic ratio of 33 separated these two groups.

Mutant B alleles All 47 subjects tested for the presence of the

CYP2D6-B allele revealed only wild-type genes, without evidence of B mutation. The distribution of the dextromethorphan metabolic ratios in these 47 subjects is shown in Fig. 2. The distribution was similar to that for total study subjects.

C/T,,, polymorphism The distributions of log[MR] values in patients

with homozygous C,,, (n = 19), heterozygous CiTlss (n = 19), and homozygous T,,, (n = 23) are shown in the upper, middle, and lower panels of Fig.

Hou et al. 413

25 1

20

2 c

*s 15

2

z 10 .

ii 5

0 !

25 i I

20 - (I)

E

.!i 15-

2

b 107

1 with n quinidine n-9

y ,,,.,,,,, Tzzy

-1.0 -0.5 0 0.5 1.0 1.5 2.0 2.5 3.0

LogMIl

Fig. 1. Distribution of logarithms of dextromethorphan metabolic ratios (log[MR]) determined before hemodialysis in 100 patients without pretreatment with quinidine sulfate (upper panel). Nine were studied again on a separate occasion after pretreatment with quinidine sulfate. The distribution of log[MR] is shown in the lower panel.

3, respectively. The subjects with homozygous C,,, had significantly lower log[MR] values (0.023 ? 0.34) than those with homozygous T,,, (0.47 t 0.33, p = 0.0067).

DISCUSSION In this study, 30 mg of dextromethorphan hydro-

bromide was given to the patients with renal failure who were receiving hemodialysis to determine their CYP2D6 phenotype by salivary analysis. The results showed that phenotyping was feasible and that only a few patients suffered from side effects. In our previous study,’ 50 mg dextromethorphan hydrobromide caused discomfort in half of Chinese normal

414 Hou et al. CLINICAL PHARMACOLOGY & THERAPEUTICS

APRIL 1996

Fig. 2. Distribution of the salivary dextromethorphan metabolic ratios (log[MR]) in patients tested for the presence of mutant B-alleles in CYP2D6 genes (hatched bars). The distribution was plotted against the distribution of salivary dextromethorphan log[MR] values in the total population of study subjects. All the 47 tested patients did not show presence of mutant B alleles.

Table I. Influence of pretreatment with quinidine on dextromethorphan metabolic ratios determined in saliva

Patient NO.

Without quinidine With quinidine

DM (nglml) DT (ngiml) MR log[MR] DM (nglml) DT (nglml) MR

1 37.7 9.2 4.1 0.61 701 <l >701 >2.85 2 10.8 2.0 5.4 0.73 197 Cl >197 >2.29 3 4.5 9.7 0.46 -0.34 56 <1 >56 >1.75 4 14.9 6.7 2.2 0.34 83 1.09 76 1.88 5 46.5 8.6 5.4 0.73 106 <1 >106 >2.03 6 21.4 7.5 2.9 0.46 93 2.56 36.5 1.56 7 2.1 4.7 0.44 -0.36 37 <l >37 B1.57 8 23.6 2.2 10.7 1.03 144 <l >144 >2.16 9 42.4 7.3 5.8 0.76 170 <l >170 ~2.23

Mean 2 SD 22.7 ? 16.3 6.4 + 2.9 4.15 ? 3.19 - 176 -c 203 1.18 +- 0.52 169.3 k 207.6* -

DM, Dextromethorphan; DT, dextrorphan; MR, metabolic ratio. For calculation of mean values the concentrations <l n&11 were assumed to be 1 ngiml, and the values with the “>” marks were assumed to be equal to the values.

*p < 0.05 compared with the MR values obtained without quinidine pretreatment.

subjects. For Chinese patients with renal failure who required hemodialysis, 30 mg dextromethorphan hydrobromide was adequate for determination of CYP2D6 phenotype.

The distribution of dextromethorphan log[MR] values in 100 Chinese patients with renal failure who

were receiving hemodialysis appeared to be normal (Fig. 1, upper panel). It is consistent with previous reports in normal Chinese subjects’8-20 in which 0% to 3% of the subjects were CYP2D6 poor metabolizers. In the present study, we administered 400 mg quinidine sulfate daily for 7 days to nine patients in

CLINICAL PHARMACOLOGY & THEBAPEUTICS VOLUME 59, NUMBER4

an attempt to convert an extensive metabolizer to a poor metabolizer. In previous studies,“,‘4 lower doses (50 mg or 150 mg) of quinidine sulfate were shown to markedly increase the metabolic ratios. However, some of the tested subjects were still extensive metabolizers despite quinidine treatment. Furthermore, our use of quinidine sulfate at such a dose was justified by the study performed by Brinn et a1.13 They pretreated seven extensive metabolizers of sparteine (one substrate of CYP2D6 enzyme) with quinidine sulfate at a daily dose of 600 to 800 mg for 7 days, and all of the subjects were converted to poor metabolizers. In our study, quinidine significantly and markedly increased the metabolic ratios (Table I). Furthermore, quinidine pretreatment shifts the metabolic ratios toward the right and out- side the distribution of those in Fig. 1, upperpanel. A metabolic ratio of 33 was shown to differentiate the two groups. The effect of quinidine on the saliva dextromethorphan metabolic ratios supports the validity of salivary analysis for determination of CYP2D6 phenotype. We have shown that salivary analysis can substitute urinary analysis for determination of dextromethorphan metabolic phenotype in normal volunteers.” An antimode of 14.0 (salivary metabolic ratio) was chosen to identify the poor metabolizers. In this study, we extended our obser- vation to include patients with renal failure. The difference in the antimodes for identification of poor metabolizers in normal subjects (14.0) and in patients with renal failure (33) may result from the different means of determination or from possible changes in pharmacokinetics in patients with renal failure (such as protein-binding, transfer of dextromethorphan or dextrorphan from blood to saliva). The actual mechanism responsible for the difference requires further study.

Genotyping was established as a reliable test for determination of CYP2D6 metabolic capability in white subjects.1,21-23 Restriction fragment length polymorphism (RFLP) with the use of Xba I identified 29 kb DNA fragments in most extensive metabolizers, whereas about half of poor metabolizers had 44 kb or 11.5 kb fragments. Allele-specific PCR analysis has shown that most of the white poor metabolizers are homozygous or heterozygous for mutant CYP2D6 alleles of the B type.1’,22,23 Dahl et a1.23 correlated CYP2D6 phenotypes with genotypes in white subjects with the use of RFLP pattern and allele-specific PCR.23 Those who were homozygous for CYP2D6 wild-type (29wt) genes were extensive metabolizers, whereas all subjects who were ho-

Hou et al. 415

15

$ 12- C/C

s n=lS

‘S i! 9- 7

6 6-

1 -1.0 -0.5 0 0.5 1.0 1.5 2.0

LogMRl

15 I

J 3 12-

C/T n-19

-1.0 -0.5 0 0.5 1.0 1.5 2.0

LogtMRl

T/T n=23

-1.0 -0.5 0 0.5 1.0 1.5 2.0

LoghlRl

Fig. 3. Distributions of log[MR] values in patients with homozygous C,,, (upper panel), heterozygous C/T,,, (middle panel), and homozygous TISR (lower panel) are shown by the solid bars. The open bars indicate the distribution of log[MR] valuess for all 61 subjects.

mozygous for CYP2D6-B (29B) genes were poor metabolizers. Furthermore, those with heterozygous 29B/29wt genes had metabolic ratios in between those of the extensive and poor metabolizers. Thus the absence of CYP2D6-B gene by PCR-based anal-

416 Hozz et al.

ysis in white subjects can reliably exclude the possibility of poor metabolizers.

In Chinese, the mutant genes responsible for deficient metabolism of CYP2D6 substrates have not yet been identified. Yue et a1.24 observed a dissociation between CYP2D6 phenotype and genotype as determined by Xba I RFLP in 21 Chinese extensive metabolizers. Seven subjects had RFLP patterns (44kb/44kb in six and 44kb/ll.S kb in one) that predicted the poor metabolizers in white subjects. Further study25 showed that the apparent 44 kb alleles from Chinese subjects were actually 40 kb in size. Wang et a1.i’ identified two poor metabolizers of debrisoquin in Taiwan. They had RFLP patterns of 29/11.5 kb and 29/16 kb, respectively. Neither of them showed mutant genes 29A, 29B, or 29C. These Chinese poor metabolizers of debrisoquin appeared to have unidentified mutant genes that resulted in a 29 kb RFLP fragment. In the present study, 47 Chinese inhabitants of Taiwan did not show a CYP2D6-B mutation. With reference to the genotypic findings in white subjects, they were most likely extensive metabolizers. This point further supports the proposition that the 100 subjects in the present study were extensive metabolizers, although we can- not exclude the presence of unidentified mutant CYP2D6 genes in these Chinese patients.

There is a wide variability in the debrisoquin or dextromethorphan metabolic ratios in the extensive metabolizers.6,8,9,18,19,26 Turgeon et a1.27 found that phenotypic debrisoquin 4-hydroxylase activity among extensive metabolizers was unrelated to the genotype as determined by the tia I RFLP. Wang et al.” identified C/T,,, polymorphism in exon 1 of CYP2D6 genes in Chinese subjects. Among extensive metabolizers, the subjects with Tis8 had higher metabolic ratios of debrisoquin. In the present study, those with T/Tis8 had higher log[MR] values of dextromethorphan in saliva than those with c/cI88* Thus, in Chinese patients with renal failure who were receiving hemodialysis, C!Tiss genotypic polymorphism accounts for, at least in part, the variability in salivary dextromethorphan metabolic ratios. This correlation between genotype and phenotype is consistent with previous findings.” Most (at least 97%) of the Chinese patients are not poor metabolizers, and determination of dextromethorphan metabolic ratios in saliva can detect the CYP2D6-type metabolic capability in extensive or intermediate metabolizers. In Chinese subjects, the latter can be predicted reasonably well by genotyping for the C/T,,, polymorphism. The clinical signif-

CLINICALPHARMACoLOGY&THERAPEUTICS APRIL 1996

icance of CYP2D6 phenotypic variability among subjects who are not poor metabolizers remains to be determined.

Identification of poor metabolizers for the CYP2D6 polymorphism is particularly important in patients with renal failure. A compound eliminated by both the liver (through metabolism by the CYP2D6 enzyme) and the kidney in similar propor- tions may not cause obvious adverse reactions in subjects with normal renal function (even in poor metabolizers). When the renal function deterio- rates, the poor metabolizers have a marked decrease in elimination of the compound, resulting in poten- tial adverse reactions. Mikus et a1.26 have demon- strated this phenomenon for flecainide. The results of the present study validate a method for determination of CYP2D6 metabolic ratios in saliva of patients with renal failure who are receiving hemodialysis.

The authors are grateful to Dr. Susan S. J. Lee and Professor Jin-Ding Huang for help in the preparation and critical review of the manuscript, respectively.

References 1.

2.

3.

4.

5.

6.

7.

8.

Gonzalez FJ, Skoda RC, Kimura S, et al. Character- ization of the common genetic defect in humans deficient in debrisoquine metabolism. Nature 1988;331: 442-6. Eichelbaum M, Gross AS. The genetic polymorphism of debrisoquine/sparteine metabolism-clinical as- pects. Pharmacol Ther 1990;46:377-94. Wang T, Roden DM, Wolfenden HT, Woosley RL, Wood AJ, Wilkinson GR. Influence of genetic polymorphism on the metabolism and disposition of encainide in man. J Pharmacol Exp Ther 1984;228:605- 11. Woosley RL, Roden DM, Dai G, et al. Co-inheritance of the polymorphic metabolism of encainide and debrisoquin. CLIN PHARMACOL THER 1986;39:282-7. Siddoway LA, Thompson KA, McAllister B, et al. Polymorphism of propafenone metabolism and disposition in man: clinical and pharmacokinetic conse- quences. Circulation 1987;75:785-91. Schmid B, Bircher J, Preisig R, Kupfer A. Polymor- phic dextromethorphan metabolism: cosegregation of oxidative 0-demethylation with debrisoquin hydroxylation. CLIN PHARMACOL THER 1985;38:618- 24. Henthorn TK, Benitez J, Avram MJ, et al. Assess- ment of the debrisoquin and dextromethorphan phenotyping tests by gaussian mixture distributions analysis. CLIN PHARMACOL THER 1989;45:328-33. Larrey D, Babany G, Tine1 M, et al. Effect of liver disease on dextromethorphan oxidation capacity and

CLINICALPHAIL~COLOC;Y&THERAPEUTIcS VOLUME 59.NUMRER4 Hou et al. 417

phenotype: a study in 107 patients. Br J Clin Pharma- co1 1989;28:297-304.

9. Hou ZY, Pickle LM, Meyer PS, Woosley RL. Salivary analysis for determination of dextromethorphan metabolic phenotype. CLIN PHARMACOL THER 1991;49:410-9.

10. Otton SV, Inaba T, Kalow W. Competitive inhibition of sparteine oxidation in human liver by B-adrenoceptor antagonists and other cardiovascular drugs. Life Sci 1983;34:73- 80.

11. Speirs CJ, Murray S, Boobis AR, Seddon CE, Davies DS. Quinidine and the identification of drugs whose elimination is impaired in subjects classified as poor metabolizers of debrisoquine. Br J Clin Pharmacol 1986;22:739-43.

12. Inaba T, Tyndale RE, Mahon WA. Quinidine: potent inhibition of sparteine and debrisoquine oxidation in vivo [Letter]. Br J Clin Pharmacol 1986;22:199-200.

13. Brinn R, Brosen K, Gram LF, Haghfelt T, Otton SV. Sparteine oxidation is practically abolished in quinidine-treated patients. Br J Clin Pharmacol 1986; 22: 194-7.

14. Zhang Y, Britto MR, Valderhaug KL, Wedlund PJ, Smith RA. Dextromethorphan: enhancing its systemic availability by way of low-dose quinidine-mediated inhibition of cytochrome P4502D6. CLIN PHARMACOL

THER 1992;51:647-55. 15. Buchert ET, Woosley RL, Swain SM, et al. Relation-

ship of CYP2D6 (debrisoquine hydroxylase) genotype to breast cancer susceptibility. Pharmacogenetics 1993;3:322-7.

16. Walsh SP, Metzger DA, Higuchi R. Chelex 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material. Biotechniques 1991;lO: 506-13.

17. Hanioka N, Kimura S, Meyer UA, Gonzalez FJ. The human CYP2D locus associated with a common genetic defect in drug oxidation: a G,,,,+A base change in intron 3 of a mutant CYP2D6 allele results in an aberrant 3’ splice recognition site. Am J Hum Genet 1990;47:994-1001.

18. Wang SL, Huang JD, Lai MD, Liu BH, Lai ML. Molecular basis of genetic variation in debrisoquin

hydroxylation in Chinese subjects: polymorphism in RFLP and DNA sequence of CYP2D6. CLIN PHARMA- COL THER 1993;53:410-8.

19. Lou YC, Ying L, Bertilsson L, Sjoqvist F. Low fre- quency of slow debrisoquine hydroxylation in a native Chinese population. Lancet 1987;2:852-3.

20. Horai Y, Nakano M, Ishizaki T, et al. Metoprolol and mephenytoin oxidation polymorphisms in Far Eastern Oriental subjects: Japanese versus mainland Chinese. CLIN PHARMACOL THER 1989;46:198-207.

21. Skoda RC, Gonzalez FJ, Demierre A, Meyer UA. Two mutant alleles of the human cytochrome P450dbl gene (P450IIDl) associated with genetically deficient metabolism of debrisoquine and other drugs. Proc Nat1 Acad Sci USA 1988;85:5240-3.

22. Heim M, Meyer UA. Genotyping of poor metabolizers of debrisoquine by allele-specific PCR amplifica- tion. Lancet 1990;336;529-32.

23. Dahl ML, Johansson I, Palmertz MP, Ingelman- Sundberg M, Sjijqvist F. Analysis of the CYP2D6 gene in relation to debrisoquin and desipramine hydroxylation in a Swedish population. CLIN PHARMACOL THER 1992;51:12-7.

24. Yue QY, Bertilsson L, Dahl-Puustinen ML, et al. Dissociation between debrisoquine hydroxylation phenotype and genotype among Chinese [Letter]. Lancet 1989;2:870.

25. Mura C, Broyard JP, Jacqz-Aigrain E, Krishnamoor- thy R. Distinct phenotypes and genotypes of debrisoquine hydroxylation among Europeans and Chinese [Letter]. Br J Clin Pharmacol 1991;32:135-6.

26. Mikus G, Gross AS, Beckmann J, Hertrampf R, Gundert-Remy U, Eichelbaum M. The influence of the sparteineldebrisoquin phenotype on the disposition of flecainide. CLIN PHARMACOL THER 1989;45: 562-7.

27. Turgeon J, Evans WE, Relling MV, Wilkinson GR, Roden DM. Phenotypic debrisoquine 4-hydroxylase activity among extensive metabolizers is unrelated to genotype as determined by the Xba I restriction fragment length polymorphism. Br J Clin Pharmacol 1991;32:283-8.

Documents

Determination of dextromethorphan metabolic phenotype by salivary analysis with a reference to genotype in Chinese patients receiving renal hemodialysis*