The predictive value of MDR1, CYP2C9, and CYP2C19 polymorphisms for phenytoin plasma levels


Phenytoin, an anticonvulsant, exhibits nonlinear pharmacokinetics with large interindividual differences. Because of its small therapeutic range with the risk of therapeutic failure or adverse drug effects in susceptible persons, therapeutic drug monitoring is frequently applied. The interindividual differences in dose response can partially be explained by known genetic polymorphisms in the metabolic enzyme CYP2C9 but a large deal of individual variability remains still unexplained. Part of this variability might be accounted for by variable uptake of phenytoin, which is a substrate of p-glycoprotein, encoded by the human MDR1 gene. We evaluated, whether phenytoin plasma levels correlate with a polymorphism in the MDR1 gene, C3435T, which is associated with intestinal PGP activity. Genotyping and analyses of plasma levels of phenytoin and metabolites in 96 healthy Turkish volunteers showed that the MDR1C > T3435 polymorphism affects phenytoin plasma levels (P = 0.064) and the metabolic ratio of p-HPPH vs phenytoin (MDR1*TT genotype, P = 0.026). The MDR1*CC genotype is more common in volunteers with low phenytoin levels (P ≤ 0.001, χ2 test). A combined analysis of variable alleles of CYP2C9, 2C19 and MDR1 revealed that the number of mutant CYP2C9 alleles is a major determinant, the number of MDR1*T alleles further contributes to the prediction of phenytoin plasma levels and CYP2C19*2 does not explain individual variability. The regression equation that fitted the data best included the number of mutant CYP2C9 and MDR*T alleles as predictory variables and explained 15.4% of the variability of phenytoin data (r2 = 0.154, P = 0.0002). Furthermore, analysis of CYP2C9 and MDR1 genotypes in 35 phenytoin-treated patients recruited from therapeutic drug monitoring showed that combined CYP2C9 and MDR1 analysis has some predictive value not only in the controlled settings of a clinical trial, but also in the daily clinical practice.


The discovery of genetic polymorphisms in drug metabolism has contributed significantly to the understanding of the inter-individual variability in dose-concentration relationships and drug response. The anticonvulsant phenytoin exhibits a nonlinear pharmacokinetics with large interindividual differences and a small therapeutic range. The use of standard dosage implies the danger of therapeutic delay or adverse drug effects in susceptible persons and demands a tedious dose finding by monitoring of plasma levels and side effects. An individualized drug therapy using pharmacogenetic characterization prior to the first dose appears therefore as a most promising tool to reduce adverse effects and to increase efficacy of pharmacotherapy.

The metabolism of phenytoin was demonstrated to be polymorphic.1 Phenytoin is predominantly metabolized by polymorphic cytochrome P450 CYP2C9 but a minor contribution of CYP2C19 has also been described.1,2 At least two protein variants (Arg144Cys and lle359Leu) affect 2C9-mediated metabolism of phenytoin3 and several genetic variants of 2C19 have been reported4,5,6 with the *2 allele (m1; a splice site mutation) causing the poor metabolizer phenotype in approximately 10% of Caucasians.5 Population studies have shown that genetic polymorphisms of both CYP2C genes play a important role in the pharmacokinetic variability of phenytoin and that defective 2C9 alleles would increase blood phenytoin levels dramatically and even patients with 2C19 mutations were suggested for careful monitoring at higher daily doses of phenytoin.7,8 However, the variability of phenytoin pharmacokinetics cannot be explained completely by CYP2C polymorphisms. So far unknown rare polymorphisms of CYP2Cs, or of other metabolizing enzymes or of factors such as the drug-transporting p-glycoprotein (pGP) might play a role. pGP is encoded by the multidrug resistance gene MDR1 which was originally identified as a gene that confers multidrug resistance.9 MDR1 functions as a cellular efflux pump and mediates the ATP-dependent transport of many drugs among them digoxin, cyclosporin, protease inhibitors, and phenytoin, through membranes.10 pGP is expressed in many tissues such as intestine, liver, and kidney, where it, considering its apical localization, prevents uptake and facilitates elimination of substrates11,12,13 and pGP is an integral part of the blood–brain barrier and protects the fetus at the placenta.14 Consequently, MDR1 serves as a defence mechanism, determines CNS concentration of substrate drugs, and influences tissue distribution and pharmacokinetics by affecting bioavailability (via intestinal absorption) or elimination (controlling intracellular concentration of parent compound for metabolic enzymes).15 One single nucleotide polymorphism in the MDR1 gene (C>T3435) has recently been identified to reduce intestinal pGP expression and to increase intestinal uptake of the pGP substrate digoxin.16

Here we show that combination of the genetic analysis of polymorphisms in CYP2C9 with that in MDR1 has predictive value for phenytoin blood concentrations.


Volunteer Study

Ninety-six volunteers (67 male and 29 female, aged 18–62, mean 26 years) of Turkish (South-East Anatolia) origin were included in the study after giving informed written consent for the study. The study protocol was approved by the Gaziantep University committee for medical research ethics as descibed earlier.17 The volunteers were proved to be healthy by medical history, medical examination, and routine laboratory parameters and they had not taken any medications prior to onset of the study.

All volunteers received a single dose of 300 mg of phenytoin and blood samples for analysis of trough levels were withdrawn after 12 h. Phenytoin and its major metabolite, 5-(para-hydroxyphenyl)-5-phenylhydantoin (p-HPPH) was analysed by HPLC as described earlier.17 Plasma was incubated with β-glucuronidase prior to analysis and HPLC did not differentiate between (R)-and (S) p-HPPH enantiomers.

Patient Study

Patients (n = 35, 29 male and six female, aged 24–82, mean 46 years) were randomly selected from the routine Therapeutic Drug Monitoring (TDM) program of the University of Vienna, Austria. All patients were of Austrian origin and received phenytoin for the treatment of epilepsy and trough levels were taken for routine control of treatment.

Since patients were recruited from therapeutic drug monitoring and not from controlled clinical trials, blood samples were taken during the daily routine and therefore at variable time points, 4–14 h after the last dose. However, peak to trough level fluctuations of phenytoin during steady state are low due to the slow oral absorption and elimination. Thus, the exact time point of blood sampling appears to be of minor influence on phenytoin plasma levels.

Since all patients had been treated with phenytoin for more than one month, plasma levels were regarded as at steady state. Daily total phenytoin doses varied from 150 to 650 mg distributed over two to three doses a day.


DNA was extracted from EDTA-treated blood using standard phenol/chloroform extraction. The CYP2C9 genotype was assessed by PCR-RFLP analysis of C>T432 mutation in exon 3 causing the exchange Arg144Cys (CYP2C9*2 allele) and exon 7 mutation (A>T1077) which codes the amino acid change ILE359VAL (CYP2C9*3 allele). The procedure was reported recently.17

The CYP2C19 allele*2 which is characterized by a splice site mutation was determined by PCR-RFLP (BamHl cleavage) as described.4

MDR1 exon 26 polymorphism (C>T3435) was detected by PCR using primers 5′GATCTGTGAACTCTTGTTTTCA3′ and 5′GAAGAGAGACTTACATTAGGC3′ with subsequent MboI cleavage as described previously.16

Statistical Analysis

The data were transformed to a log scale to normalize the distribution. The influence of MDR1 exon 26, CYP2C9, and −2C19 polymorphisms on phenytoin plasma concentrations and on the metabolic ratio of p-HPPH vs its parent compound was assessed by ANOVA using the lifelong constant genotype as independent and plasma levels as dependent variables. P-values were corrected for multiple comparisons according to Bonferroni Dunn.

Using the statistical distribution parameters, the 25th and the 75th percentile, as separator, the volunteers were divided into three classes of low (<3.79 mg l−1, n = 24, 25th percentile), high (>5.79 mg l−1, n = 24, 75th percentile), and medium (3.79–5.79 mg l−1, n = 48, interquartile range) plasma concentrations to contemplate the impact of the MDR1 genotype. The chi-square test was used to demonstrate equality of distribution of MDR1 genotypes within these classes.

The quantitative impact of genotypes to explain the variability in phenytoin plasma concentration and on the metabolic ratio of p-HPPH/phenytoin was calculated by multiple linear regression analysis. CYP2C9 genotypes were aggregated into one combined variable, whereas MDR1 exon 26 and CYP2C19 genotypes were transformed each into a variable. These variables were ordinally scaled reflecting the number of mutant (defective) alleles. The CYP2C9*2 and −*3, the CYP2C19*2, and the MDR1*T alleles were assumed to possess a reduced functional capacity. The equation that fitted the data best was determined by building regression models, considering all possible subsets of predictor variables (CYP2C9, CYP2C19, MDR1 genotypes, and age). The model with the subset that gave the largest value for the corrected r2 was selected to explain the variability in phenytoin plasma levels.

In patients from therapeutic drug monitoring, dose-corrected phenytoin plasma levels were used by dividing phenytoin plasma levels by total daily dose of phenytoin to minimize dose-related variation of phenytoin plasma levels. Since phenytoin exhibits non-linear pharmacokinetics, dose-correction can not completely eliminate dose-related effects on plasma levels. The correlation between genotypes and dose-corrected phenytoin plasma levels in patients from therapeutic drug monitoring was analyzed using the Jonckheere–Terpstra test, a nonparametric statistical procedure that considers the quantitative relationship (number of defective alleles) between the levels of a variable. Statistical significance was set at P < 0.05 (two-sided); 0.05 < P < 0.10 was regarded as statistical trend. Statistical analysis was performed with the SPSS 10.0 package (SPSS Inc, Chicago, USA).


The gene frequencies of CYP2C9, CYP2C19, and MDR1C>T3435 and mean (SD) phenytoin plasma levels are given in Table 1. Mean (SD) trough levels taken 12 h after a 300-mg test dose of phenytoin were 4.68 (1.48) mg l−1 and mean (SD) 12 h plasma concentration of the major metabolite p-HPPH was 1.64 (0.66) mg l−1.

Table 1 Phenytoin plasma concentration and frequency of CYP2C9, CYP2C19, and MDR1C > T3435 genotypes in 96 healthy Turkish volunteers

Influence of CYP2C9 and CYP2C19

The frequency distribution of phenytoin plasma levels and metabolic phenytoin/p-HPPH ratios within the genetically defined CYP2C9 subgroups is illustrated in Figure 1.

Figure 1

Frequency distribution of (a) phenytoin trough levels and (b) the metabolic ratio (p-HPPH/phenytoin) in 96 healthy volunteers with respect to the number of mutant CYP2C9*2 and *3 alleles.

The phenytoin plasma levels correlated statistically significantly with the CYP2C9 genotype (P < 0.0001, Anova) and increased with the number of cysteine144 and leucin359 alleles (mean/SD: 4.20/1.25, 5.53/1.42, and 6.41/1.66 mg l−1 in CYP2C9*1/1, *1/2 plus *1/3, and *2/2 plus *3/3, respectively). Phenytoin plasma levels in volunteers homozygous for the wild-type allele (genotype: CYP2C9*1/1) were statistically significantly different from volunteers with one mutant allele (genotypes CYP2C9*1/2 and *1/3; P = 0.0016, ANOVA with Bonferroni Dunn correction) and from volunteers with two mutant alleles (genotypes CYP2C9*2/2 and *3/3; P = 0.049, ANOVA with Bonferroni Dunn correction). Unlike plasma levels of the parent compound, the metabolic phenytoin ratio (p-HPPH/phenytoin) is less dependent on bioavailability and volume of distribution and could therefore be regarded to be a better measure for CYP2C9 phenotype (Figure 1b). The metabolic ratio decreased with increasing number of defective 2C9 alleles (P < 0.001) reflecting the lower metabolic activity of the mutant alleles (mean/SD: 1.83/0.67, 1.34/0.38, 0.76/0.48 in CYP2C9*1/1, *1/2 plus *1/3, and *2/2 plus *3/3, respectively).

In contrast to CYP2C9 alleles, the poor metabolizer-associated CYP2C19*2 allele (the CYP2C19*3 was not found in this sample) failed to show any association either to phenytoin plasma levels or to the metabolic ratio p-HPPH/phenytoin.

Influence of MDR1

Although the hereditary MDR1C>T3435 polymorphism was only of minor impact, it still trended to affect phenytoin plasma levels (P = 0.064, Figure 2a) and the MDR1*TT genotype influenced statistically significantly the metabolic ratio of p-HPPH vs phenytoin (P = 0.026, Figure 2b). Figure 2 indicates that MDR1 genotypes are not equally distributed and that the MDR1*CC genotype is more common in volunteers with low phenytoin levels.

Figure 2

Frequency distribution of (a) phenytoin trough levels and (b) the metabolic ratio (p-HPPH/phenytoin) in 96 healthy volunteers with respect to the MDR1C>T3435 genotype.

Based on their frequency distribution, the phenytoin levels were divided into three classes of low (<3.79 mg l−1, n = 24, 25th percentile) medium (3.79–5.79 mg l−1, n = 48, interquartile range), and high (>5.79 mg l−1, n = 24, 75th percentile) plasma concentrations. The resulting histogram (Figure 3), grouped according to the MDR1 genotype, demonstrates clearly the significant overrepresentation of the MDR1*CC genotype in low phenytoin plasma concentrations (P ≤ 0.001, χ2-test).

Figure 3

Frequency distribution of the MDR1C>T3435 genotype in subgroups of healthy volunteers with low (<3.79 mg l−1, n = 24), medium (3.79–5.79 mg l−1, n = 48) and high (>5.79 mg l−1, n = 24) phenytoin plasma levels. The MDR1*CC genotype was statistically significantly overrepresented in the lower group (P ≤ 0.001, χ2 test).

Combined Analysis

To reveal the individual contribution of the investigated polymorphisms in explaining the variability of phenytoin plasma levels, multiple regression analysis was performed. Several prediction equations were computed and the goodness of fit of each model was estimated using the corrected r2. It appeared that CYP2C19*2 did not contribute, the number of mutant CYP2C9 alleles was the major explanatory factor (14.1%), and the number of MDR1*T alleles has some additional impact (1.3%) and improved the prediction of variability of phenytoin plasma levels by 9.2% to 15.4%. Thus, the regression equation that fitted the data best included the number of mutant CYP2C9 and MDR*T alleles as predicting variables (r2 = 0.154, P = 0.0002).

Therapeutic Drug Monitoring

To evaluate the value of MDR1 and Cyp2C9 genotyping for routine use in clinical practice, we investigated 35 patients recruited from TDM. The results in Figure 4 show that the consequences from our analysis can be applied to and do reflect the situation in patients who received phenytoin for therapy. The dose-corrected plasma phenytoin levels increased with both the number of mutant Cyp2C9 alleles (P = 0.06, Jonckheere–Terpstra test) as well as the number of MDR1*T alleles (P = 0.03, Jonckheere–Terpstra test).

Figure 4

Distribution of dose-corrected phenytoin plasma levels in 35 patients recruited from routine therapeutic drug monitoring. Bars represent the median of the respective subgroup.


The value of CYP2C9, CYP2C19, and MDR1 genotypes to predict the variability of phenytoin trough levels was studied. The genotype frequencies in healthy Turkish volunteers were calculated to allele frequencies of mutant CYP2C9*2, −*3, CYP2C19*2, and MDR13435T alleles of 8.9%, 9.9%, 11.7%, and 47.4%, respectively and were in a similar range as reported for other Caucasian populations.4,5,16,17,18,19,20

Phenytoin fulfills many criteria for TDM. It has a relative narrow therapeutic range combined with saturable kinetics resulting in nonlinear dose-concentration relationships. Patients often comply poorly with the drug regimen and phenytoin pharmacokinetics exhibits large intersubject differences implying the existence of subpopulations of patients with vastly different dose-requirements. The dramatic consequences are described in a case report: toxic symptoms caused by excessive phenytoin levels (32.6 mg l−1) at a ‘lower than standard’ dosage were reported in a patient carrying the CYP2C9*3 and the CYP2C19*2 alleles.21 Conventional TDM applies the measurement of plasma levels after the start of treatment with a standard dose implying detection of toxicity after risk exposure or causing a delay of treatment in nonresponders. The goal of pharmacogenetic research is to develop a ‘susceptibility profile’ for identifying individuals at risk prior to the first dosage.22,23

The coinheritance of phenytoin metabolism with known drug hydroxylation polymorphism has long been identified1 and the polymorphic CYP2C9 has been recognized as the main responsible enzyme catalyzing the formation of p-HPPH.3,24 The impairment of hydroxylation capacity and the clinical importance of the Leu359 allele in phenytoin treatment in particular for predicting high phenytoin concentrations even at lower daily doses has been described.8,21 In the present study, Leu359 and Cys144 had approximately the same impact and were responsible for 14% of the variability of phenytoin trough levels. Although five of the seven highest phenytoin levels were explained by mutant 2C9 alleles, volunteers with at least one mutant allele had only about 30% higher plasma levels than homozygous wildtype volunteers and a major part of variability remained unexplained. CYP2C19 has been described as contributing to the metabolism of phenytoin, and patients with CYP2C19 mutations have been suggested to be endangered at higher dosage.7 However, CYP2C19 generates (R) p-HPPH, which accounts for only 2–10% of total p-HPPH.1,25 In our study, which is in agreement with these findings, CYP2C19 alleles were not found to influence phenytoin plasma levels. MDR1 polymorphism was studied as another candidate to improve the prediction of variability of phenytoin plasma levels. MDR1 encodes pGP and pGP-mediated increased drug efflux that removes drugs from the cell before exerting cytotoxic effects was first observed in tumors to cause resistance against chemotherapeutic agents.9,13,26 MDR1 is expressed in various tissues including intestine, liver and kidneys and variable expression has been correlated to differences in drug uptake and response27,28,29 Furthermore, pGP is abundant in the blood–brain barrier and conciliates the penetration of substrate drugs into the brain.30 This might confer central nervous side effects or therapeutic outcome. A systematic screen of the entire MDR1 gene for polymorphisms was performed by Hoffmeyer et al. All 28 exons including the core promoter region and exon-intron boundaries that are important for mRNA splicing, were sequenced from the genomic DNA from healthy Caucasians and 15 single nucleotide polymorphisms have been described, among them six in the coding region of which three lead to an alteration of the protein. One of these polymorphisms, even though in wobble position, was demonstrated to be of functional consequence: the MDR1T3435 allele reduced intestinal pGP expression and increased digoxin plasma levels significantly.16 This SNP is located at a noncoding non-promoter position in the MDR1 gene and is unlikely to influence pGP expression. It is more likely linked to other so far unidentified changes in regions of the MDR1 gene that control expression, eg in the promoter or enhancer region, or in sequences that are important for mRNA processing. To evaluate whether the MDR1C>T3435 polymorphism has clinical importance in other PGP-transported drugs, the present study on the pharmacokinetics of phenytoin was initiated: although an influence of the MDR1 genotype on phenytoin plasma level, explainable by pGP-mediated variable drug uptake, was observed, the effect was less pronounced compared to that observed by Hoffmeyer et al on digoxin. However, the MDR1C>T3435 polymorphism influenced statistically significantly the metabolic ratio p-HPPH/phenytoin and improved the respective regression model by almost 20% (from 22% for CYP2C9 to 26% for CYP2C9 and MDR1). Because the metabolic ratio is less confounded by factors such as bioavailability or volume of distribution, it is an important pharmacological measure for emphasizing the role of MDR1 in phenytoin pharmacokinetics. Our data indicate that for overall phenytoin levels, MDR1 genotyping might be of limited predictive value. However, the high pGP-expressing MDR13435C allele is significantly overrepresented in volunteers with low phenytoin levels (Figure 3) and might be therefore, like CYP2C19*2 and *3 alleles for toxic concentrations, predictive for subtherapeutic plasma levels. Such low levels put patients at risk of encountering therapeutic failure. This is in agreement with previous suggestions that MDR1 gene expression and persistently low phenytoin share a common pathway liable to induce drug-resistant epilepsy.31

One major question in most analyses of genotype–phenotype associations is whether the results that were obtained in clinical trials under controlled conditions can also be applied in the daily clinical practice. To evaluate the value of MDR1 and CYP2C9 genotyping for clinical practice, we investigated 35 patients recruited from routine therapeutic drug monitoring and could demonstrate that the consequences from our analysis are indeed applicable to patients treated with phenytoin. Phenytoin plasma levels (dose-corrected) increased with both the number of mutant CYP2C9 alleles (P = 0.06, Jonckheere–Terpstra test) as well as with the number of MDR1*T alleles (P = 0.03, Jonckheere–Terpstra test). One concern with our clinical data is that blood sampling for the therapeutic drug monitoring was performed during the daily clinical routine and therefore at variable timepoints (4–14 h after the last dose). However, peak to trough fluctuations of phenytoin plasma levels during steady state are low due to the slow oral absorption and elimination. Thus, the exact time point of blood sampling appears to be of minor influence on phenytoin plasma levels.

Applying CYP2C9 and MDR1 genotyping should help to predict outliers (ie those who will have particular high or low plasma levels after standard dosage) prior to onset of therapy. This is particularly important in phenytoin therapy, because the narrow therapeutic range enhances the risk of achieving toxic concentrations and on the other hand, therapeutic failure can have dramatic consequences by provoking seizures.

One very important question is whether individual variability of CYP2C9 and MDR1 affects therapeutic response to phenytoin. Drug response depends on drug levels and is additionally influenced by many other endogenous and exogenous factors (receptors, neurotransmitters, disease state). Therefore, large studies with extended patient numbers are required to obtain sufficient data that permit thorough statistical analyses. Our TDM data are the results from clinical testing which is ongoing, and which should enable us to address this question in future. Moreover, unlike phenotyping methods, DNA-based tests can be used during active drug therapy to predict dosage requirements or to understand mechanisms involved in extremely low or high drug levels. This carries a great potential as a complement to traditional TDM.


R Kerb, R Schlagenhaufer and U Brinkmann are employed at EPIDAUROS, a Pharmacogenetics Company which has applied for protection of intellectual property regarding pharmacogenetic applications of selected MDR1 polymorphisms.


  1. 1

    Fritz S, Lindner W, Roots I, Frey BM, Küpfer A . Stereochemistry of aromatic phenytoin hydroxylation in various drug hydroxylation phenotypes in humans J Pharmacol Exp Ther 1987 241: 615–622

  2. 2

    Yasumori T, Chen LS, Li QH, Ueda M, Tsuzuki T, Goldstein JA et al. Human CYP2C-mediated stereoselective phenytoin hydroxylation in Japanese: difference in chiral preference of CYP2C9 and CYP2C19 Biochem Pharmacol 1999 57: 1297–1303

  3. 3

    Veronese ME, Doecke CJ, Mackenzie PI, McManus ME, Miners JO, Rees DL et al. Site-directed mutation studies of human liver cytochrome P-450 isoenzymes in the CYP2C subfamily Biochem J 1993 289: 533–538

  4. 4

    De Morais SM, Wilkinson GR, Blaisdell J, Meyer UA, Nakamura K, Goldstein JA . Identification of a new genetic defect responsible for the polymorphism of (S)-mephenytoin metabolism in Japanese Mol Pharmacol 1994 46: 594–598

  5. 5

    De Morais SM, Wilkinson GR, Blaisdell J, Nakamura K, Meyer UA, Goldstein JA . The major genetic defect responsible for the polymorphism of S-mephenytoin metabolism in humans J Biol Chem 1994 269: 15419–15422

  6. 6

    Ferguson RJ, De Morais SM, Benhamou S, Bouchardy C, Blaisdell J, Ibeanu G et al. A new genetic defect in human CYP2C19: mutation of the initiation codon is responsible for poor metabolism of S-mephenytoin J Pharmacol Exp Ther 1998 284: 356–361

  7. 7

    Mamiya K, Ieiri I, Shimamoto J, Yukawa E, Imai J, Ninomiya H, Yamada H et al. The effects of genetic polymorphisms of CYP2C9 and CYP2C19 on phenytoin metabolism in Japanese adult patients with epilepsy: studies in stereoselective hydroxylation and population pharmacokinetics Epilepsia 1998 39: 1317–1323

  8. 8

    Odani A, Hashimoto Y, Otsuki Y, Uwai Y, Hattori H, Furusho K et al. Genetic polymorphism of the CYP2C subfamily and its effect on the pharmacokinetics of phenytoin in Japanese patients with epilepsy Clin Pharmacol Ther 1997 62: 287–292

  9. 9

    Juliano RL, Ling V . A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants Biochim Biophys Acta 1976 455: 152–162

  10. 10

    Seelig A . A general pattern for substrate recognition by P-glycoprotein Eur J Biochem 1998 251: 252–261

  11. 11

    Ambudkar SV, Dey S, Hrycyna CA, Ramachandra M, Pastan I, Gottesman MM . Biochemical, cellular, and pharmacological aspects of the multidrug transporter Annu Rev Pharmacol Toxicol 1999 39: 361–398

  12. 12

    Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, Pastan I, Willingham MC . Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues Proc Natl Acad Sci U S A 1987 84: 7735–7738

  13. 13

    Ueda K, Clark DP, Chen CJ, Roninson IB, Gottesman MM, Pastan I . The human multidrug resistance (mdr1) gene. cDNA cloning and transcription initiation J Biol Chem 1987 262: 505–508

  14. 14

    Cordon-Cardo C, O'Brien JP, Boccia J, Casals D, Bertino JR, Melamed MR . Expression of the multidrug resistance gene product (P-glycoprotein) in human normal and tumor tissues J Histochem Cytochem 1990 38: 1277–1287

  15. 15

    Fromm MF . P-glycoprotein: a defense mechanism limiting oral bioavailability and CNS accumulation of drugs J Pharmacol Exp Ther 2000 293: 376–382

  16. 16

    Hoffmeyer S, Burk O, von Richter O, Arnold HP, Brockmöller J Johne A et al. Functional polymorphisms of the human multidrug-resistance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in vivo Proc Natl Acad Sci U S A 2000 97: 3473–3478

  17. 17

    Aynacioglu AS, Brockmöller J, Bauer S, Sachse C, Güzelbey P, Öngen Z et al. Frequency of cytochrome P450 CYP2C9 variants in a Turkish population and functional relevance for phenytoin Br J Clin Pharmacol 1999 48: 409–415

  18. 18

    Aynacioglu AS, Sachse C, Bozkurt A, Kortunay S, Nacak M, Schröder T et al. Low frequency of defective alleles of cytochrome P450 enzymes 2C19 and 2D6 in the Turkish population Clin Pharmacol Ther 1999 66: 185–192

  19. 19

    Stubbins MJ, Harries LW, Smith G, Tarbit MH, Wolf CR . Genetic analysis of the human cytochrome P450 CYP2C9 locus Pharmacogenetics 1996 6: 429–439

  20. 20

    Yasar U, Eliasson E, Dahl ML, Johansson I, Ingelman-Sundberg M, Sjöquist E . Validation of methods for CYP2C9 genotyping: frequencies of mutant alleles in a Swedish population Biochem Biophys Res Commun 1999 254: 628–631

  21. 21

    Ninomiya H, Mamiya K, Matsuo S, Ieiri I, Higuchi S, Tashiro N . Genetic polymorphism of the CYP2C subfamily and excessive serum phenytoin concentration with central nervous system intoxication Ther Drug Monit 2000 22: 230–232

  22. 22

    Brockmöller J, Cascorbi I, Henning S, Meisel C, Roots I . Molecular genetics of cancer susceptibility Pharmacology 2000 61: 212–227

  23. 23

    Eichelbaum M, Evert B . Influence of pharmacogenetics on drug disposition and response Clin Exp Pharmacol Physiol 1996 23: 983–985

  24. 24

    Bhasker CR, Miners JO, Coulter S, Birkett DJ . Allelic and functional variability of cytochrome P4502C9 Pharmacogenetics 1997 7: 51–58

  25. 25

    Ieiri I, Mamiya K, Urae A, Wada Y, Kimura M, Irie S et al. Stereoselective 4′ hydroxylation of phenytoin: relationship to (S)-mephenytoin polymorphism in Japanese Br J Clin Pharmacol 1997 43: 441–445

  26. 26

    Gottesman MM, Pastan I, Ambudkar SV . P-glycoprotein and multidrug resistance Curr Opin Genet Dev 1996 6: 610–617

  27. 27

    Mickley LA, Lee JS, Weng Z, Zhan Z, Alvarez M, Wilson W et al. Genetic polymorphism in MDR-1: a tool for examining allelic expression in normal cells, unselected and drug-selected cell lines, and human tumors Blood 1998 91: 1749–1756

  28. 28

    Schinkel AH, Mayer U, Wagenaar E, Mol CA, van Deemter L, Smit JJ et al. Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins Pharmacogenetics 2000 10: 583–590

  29. 29

    Schinkel AH, Wagenaar E, van Deemter L, Mol CA, Borst P . Absence of the mdr1a P-glycoprotein in mice affects tissue distribution and pharmacokinetics of dexamethasone, digoxin, and cyclosporin A J Clin Invest 1995 96: 1698–1705

  30. 30

    Schinkel AH, Wagenaar E, Mol CA, van Deemter L . P-glycoprotein in the blood-brain barrier of mice influences the brain penetration and pharmacological activity of many drugs J Clin Invest 1996 97: 2517–2524

  31. 31

    Lazarowski A, Sevlever G, Taratuto A, Massaro M, Rabinowicz A . Tuberous sclerosis associated with MDR1 gene expression and drug-resistant epilepsy Pediatr Neurol 1999 21: 731–734

Download references


Parts of this work were supported by grants No. 01EC9408 (Clinical Pharmacogenetics), 01GG9845, and 01GG9848 from the German Ministry for Education and Research.

Author information

Correspondence to R Kerb or U Brinkmann.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kerb, R., Aynacioglu, A., Brockmöller, J. et al. The predictive value of MDR1, CYP2C9, and CYP2C19 polymorphisms for phenytoin plasma levels. Pharmacogenomics J 1, 204–210 (2001).

Download citation


  • pharmacogenetics
  • polymorphism
  • TDM

Further reading