European Journal of Clinical Pharmacology

, Volume 67, Issue 8, pp 787–795 | Cite as

Tacrolimus dosing in Chinese renal transplant recipients: a population-based pharmacogenetics study

  • Liang Li
  • Chuan-Jiang Li
  • Lei Zheng
  • Yan-Jun Zhang
  • Hai-Xia Jiang
  • Bo Si-Tu
  • Zhong-Hai Li
Pharmacogenetics

Abstract

Objectives

The aims of this study were to examine the effects of genetic and clinical factors on the maintenance dose of tacrolimus in patients following renal transplantation and to develop a tacrolimus-dosing model that could be combined with associated factors.

Patients and methods

This study included 142 renal transplant recipients who received tacrolimus as immunosuppressive agent. CYP3A5, MDR1 and NR1I2 gene polymorphisms were identified based on the SNaPshot assay. The relationship between the genetic and clinical factors and tacrolimus maintenance dose as well as between dose-corrected tacrolimus concentration was examined.

Results

CYP3A5 genotype, body weight, haematocrit, haemoglobin and total bilirubin significantly influenced the maintenance tacrolimus dose. The tacrolimus-dosing model derived from linear regression model accounted for 40.5% of total variations in the tacrolimus maintenance dose.

Conclusions

A pharmacogenetics-based dosing model has been developed for the prediction of the tacrolimus maintenance dose in renal transplant recipients. This model may be useful in helping clinicians prescribe the initial tacrolimus dose with greater safety and effectiveness.

Keywords

CYP3A5  Dosing model Pharmacogenetics Renal transplantation Tacrolimus 

Introduction

Tacrolimus is an effective immunosuppressive drug widely used in solid organ transplantation to prevent acute rejection and chronic allograft dysfunction [1]. It is characterized by a narrow therapeutic index and large interindividual variability in its pharmacokinetics [2]. As a result, clinicians should monitor the concentration of tacrolimus so that its dosage can be adjusted to achieve therapeutic efficacy [3]. In clinical practice, it may take 2 or more weeks to reach the stable maintenance dose during which time organ rejection or tacrolimus toxicity may occur due to fluctuating tacrolimus levels [4]. Consequently, it is necessary to achieve the stable maintenance dose as soon as possible so that the therapeutic efficacy of tacrolimus can be improved and the adverse side effects reduced.

Interindividual variability in tacrolimus dose results from both genetic and clinical factors. Tacrolimus is a substrate of cytochrome P450 (CYP) 3A5 and P-glycoprotein (P-gp), which are encoded by the CYP3A5 and multidrug resistance-1 (MDR1) genes, respectively [3]. The different degrees of intestinal and hepatic expression and bioactivity of P-gp and the CYP enzymes regulate the absorptive barrier and the hepatic clearance of tacrolimus and, therefore, determine the interindividual variability of tacrolimus. A single nucleotide polymorphism (SNP) (6986A > G) in intron 3 of CYP3A5, referred to as CYP3A5*3, results in the absence of CYP3A5 protein (CYP3A5 nonexpressors), while the A nucleotide, referred to as CYP3A5*1, expresses large amounts of CYP3A5 protein (CYP3A5 expressors) [5]. Studies have revealed that CYP3A5*1 carriers require a higher dose of tacrolimus than CYP3A5*3 homozygotes [6, 7]. Comparably, SNPs −129T > C, 1236C > T, 2677G > T/A and 3435C > T for the MDR1 gene are associated with changed P-gp expression [8]. The association between MDR1 SNPs and concentration of tacrolimus in the blood has been investigated, with contradictory results [9]. In a number of studies, MDR1 haplotypes, which are the combination of MDR1 SNPs, have been studied in terms of tacrolimus pharmacokinetics. However, the association between MDR1 haplotypes and tacrolimus blood concentrations is also controversial [10, 11, 12]. Recently, pregnane X receptor (PXR), encoded by the NR1I2 gene, has been found to regulate the expression of a series of genes involved in the metabolism of xenobiotics, including CYP3A and MDR1 [13]. The SNP (-25385C > T) of NR1I2 has been reported to influence CYP3A or P-gp expression and be associated with tacrolimus pharmacokinetics [14]. In addition to genetic factors, clinical factors, such as haematocrit status [14], haemoglobin, albumin, total bilirubin (TBIL), unconjugated bilirubin (IBIL) and liver function [15], can be correlated with tacrolimus pharmacokinetics.

In recent years, drug dosing models that estimate the dose requirement for patients based on both genetic and clinical factors have been developed, especially in those cases involving the clinical use of the drugs with a narrow therapeutic index. Several models have been validated in different populations, and the results show that the drug dosing model is a promising guiding principle for planning safe and effective therapy in patients [16, 17]. In terms of the clinical use of tacrolimus, mathematical models have been used to predict individual pharmacokinetic parameters in organ transplant recipients [14, 18]. However, the development and application of a tacrolimus dosing model has not yet been reported. In clinical practice, tacrolimus overdosing can result in toxicity and the loss of graft function, while underdosing can result in immunological graft rejection owing to the inadequate suppressive effects. Consequently, a good tacrolimus dosing model would be very useful to resolve the clinical dilemma. We therefore conducted the study reported here to investigate the impact of SNPs for the CYP3A5, NR1I2, MDR1 and MDR1 haplotypes on the stable tacrolimus dose requirement as well as dose-corrected tacrolimus concentration and to develop a tacrolimus dosing model based on both genetic and clinical factors for estimating the dose requirement of patients.

Methods

Patient population and data collection

All information required for the study was extracted from medical records of renal transplant patients of Nanfang Hospital, Guangzhou (PR China) who were administered tacrolimus between January 2007 and May 2010. Demographic characteristics, laboratory test results and drug administration history were obtained from the electronic medical records. All patients were ≥18 years of age and gave written informed consent before being enrolled in the study. The study protocol was approved by the local ethical committee. Patients with the following diseases and/or medical conditions were excluded due to possible interference with tacrolimus pharmacokinetics and pharmacodynamics: hepatitis B, hepatitis C, cancer, systemic lupus erythematosus (SLE) on long-term hormone therapy, both liver and renal transplantation and second renal transplantation. Those patients in whom the stable maintenance tacrolimus dose could not be reached were also excluded. The stable maintenance tacrolimus dose was considered to be achievement of the dosage that fell within the predefined tacrolimus therapeutic range (10–12 ng/ml) for >2 consecutive days and for which the trough blood concentration during the following period did not deviate from the range of 9–14 ng/ml. This dosage would not change and was considered to be the stable maintenance tacrolimus dose. A total of 142 patients (including 16 diabetic patients) met the inclusion criteria and were recruited in the study. The laboratory parameters, including haemoglobin, haematocrit, albumin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), TBIL and IBIL, were obtained for the study when the patients reached the predefined tacrolimus therapeutic range.

Immunosuppressant regimens and tacrolimus measurement

All patients were treated with a combination of immunosuppressants consisting of tacrolimus, mycophenorate mofetil and steroids. None of the drugs used in the therapeutic regimens was known to interact significantly with tacrolimus pharmacokinetics, with the exception of the steroids. The standard steroid regimen was 1000 mg methylprednisolone given intravenously at the time of surgery, decreasing to 500, 250 and 250 mg on days 1, 2 and 3 post-operation, respectively. Treatment with30 mg of oral prednisolone was initiated on day 4, with a 5 mg dose reduction daily up to day 9 to reach a maintenance dose of 5 mg daily. The first tacrolimus administration was given orally approximately 12 h after the transplantation, with the initial dosage calculated according to the weight of the patient (0.10 mg/kg body weight, twice daily) and further individually adjusted according to trough blood concentrations. The target trough blood concentration was between 10 and 12 ng/ml; blood concentrations were measured using the Microparticle Enzyme ImmunoAssay on the IMx analyser (Abbott Laboratories, Chicago, IL). The trough blood concentration was dose-corrected using the concentration/dose ratio, obtained by dividing the tacrolimus trough concentration by the corresponding 24-h dose, on an milligram per kilogram basis.

SNP genotyping and haplotype inference

Genomic DNA was extracted from EDTA-treated blood using the TIANamp Genomic DNA kit (Tiangen Biotech, Beijing, China) and stored at −20°C for analysis. SNPs CYP3A5 6986A > G (rs776746, CYP3A5*3 allele), MDR1 -129T > C (rs3213619), MDR1 1236C > T (rs1128503), MDR1 2677G > T/A (rs2032582), MDR1 3435C > T (rs1045642) and NR1I2 -25385C > T (rs3814055) were analysed in this study. The detection of SNPs was based on the SNaPshot assay using an Applied Biosystems Multiplex kit (Invitrogen, Shanghai, China). MDR1 -129T > C (rs3213619), MDR1 1236C > T (rs1128503), MDR1 2677G > T/A (rs2032582) and MDR1 3435C > T (rs1045642) polymorphisms were used to infer the haplotype of the MDR1 gene using the PHASE 2.1.1 program.

Statistical analysis

All results were expressed as the mean ± standard deviation (SD). Differences were considered to be statistically significant at alpha level of 0.05. A one-way analysis of variance (ANOVA) was used to assess the statistical significance of differences in stable tacrolimus dose and the time to reach the stable dose as well as dose-corrected tacrolimus concentration among the category variables, such as gender, genotype and haplotype. The Pearson correlation was used to analyse the relation between the continuous variables and stable tacrolimus dose. A multiple stepwise regression analysis was performed using variables with statistical significance to establish the tacrolimus-dosing model. The stable tacrolimus dose was log-transformed to attain normality, and the final linear regression would then undergo exponentiation to calculate the required tacrolimus dose. To select candidate predictors for the multiple regression models, we examined each continuous variable individually against tacrolimus dose requirements based on the Pearson’s correlation test. The category variables and the continuous variables that were associated (P < 0.05) with the tacrolimus dose were considered further in the stepwise regression. The dummy variables were used for the category variables in the stepwise regression models. All statistical analyses were performed using the SPSS software package (ver. 13.0; SPSS, Chicago, IL).

Results

Patient characteristics

Between January 2007 and May 2010, a total of 228 renal transplant patients at Nanfang Hospital received tacrolimus. Of these, 86 were excluded for the following reasons: hepatitis B virus (41 patients), hepatitis C virus (4), second renal transplantation (8), both liver and renal transplantation (4), long-term hormone therapy for SLE (2), cancer (2), less than 18 years of age (1) and inability to achieve stable maintenance tacrolimus dose (24). Consequently, 142 renal transplant recipients were included in the study to develop the tacrolimus-dosing model. The demographics, clinical characteristics and genotypic distribution of all the patients are listed in Table 1. The mean tacrolimus maintenance dose among these patients was 7.80 ± 3.22 mg/day (range 3–16 mg/day). The mean time to reach the maintenance dose was 16.64 ± 6.58 days (range 3–39 days), and the mean trough tacrolimus concentration was 10.9 ± 0.9 ng/ml (range 10.0–11.9 ng/ml).
Table 1

Demographics, clinical characteristics and allelic variant distribution of the patient cohort

Variable

Model-building cohort (n =142)

Age (years), (mean ± SD)

42.6 ± 12.9

Gender (male/female)

99/43

Body weight (kg) (mean ± SD)

58.4 ± 10.0

Diabetes (yes/no)

16/126

Haematocrit (%)

0.313 ± 0.0488

Haemoglobin (g/l)

104 ± 16.5

Albumin (g/l)

40.7 ± 23.6

ALT (U/l)

36.4 ± 41.1

AST (U/l)

19.8 ± 11.9

TBIL (μmol/l)

9.76 ± 3.82

IBIL (μmol/l)

6.78 ± 2.87

Tacrolimus maintenance dose (mg/day)

7.80 ± 3.22

Time to reach the stable dose (day)

16.64 ± 6.58

Tacrolimus concentration (ng/ml)

10.9 ± 0.9

Concentration/dose ratio (ng/ml)/(mg/kg)

96.3 ± 50.0

CYP3A5 (*1/*1, *1/*3, *3/*3)

11, 65, 66

MDR1-129 (T/T, T/C, C/C)

130, 11, 1

MDR1 1236 (C/C, C/T, T/T)

13, 77, 52

MDR1 2677 (G/G, G/T, G/A, A/T, T/T, A/A)

42, 48, 17, 15, 19, 1

MDR1 3435 (C/C, C/T, T/T)

61, 61, 20

NR1I2 -25385 (C/C, C/T, T/T)

94, 41, 7

SD, Standard deviation; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TBIL, total bilirubin; IBIL, unconjugated bilirubin

Difference in the dose and the time to reach a stable dose for each category variable

We examined the association of the category variables with maintenance tacrolimus dose and the time to reach a stable dose. Of the eight category variables, only the CYP3A5 genotype presented a significant association with daily tacrolimus dose (Table 2). The tacrolimus dose was lower (P < 0.001) among patients with the CYP3A5 *3/*3 genotype (5.79 ± 2.16 mg/day) than in those with the CYP3A5*1/*3 (9.35 ± 2.86 mg/day) or CYP3A5*1/*1 (10.68 ± 3.49 mg/day) genotypes. The gender, whether or not having diabetes, and the presence of the MDR1 -129T > C, MDR1 1236C > T, MDR1 2677G > T/A, MDR1 3435C > T and NR1I2 -25385C > T SNPs were not significantly associated with daily tacrolimus dose. None of the eight category variables demonstrated a significant association with the time to reach stable tacrolimus dose.
Table 2

Differences in tacrolimus dose and the time to reach a stable dose for each category variable

Category variables

Patients (n = 142)

Stable dose (mg/day)

P value

Concentration/dose ratio(ng/ml)/(mg/kg)

value

Time to stable dose (days)

P value

Gender

  Male

99

7.96 ± 3.31

0.360

99.5 ± 50.9

0.245

16.26 ± 6.60

0.300

  Female

43

7.42 ± 3.02

 

88.8 ± 47.5

 

17.51 ± 6.52

 

Diabetes

  Yes

16

7.72 ± 3.39

0.920

112.0 ± 54.6

0.183

16.63 ± 4.62

0.992

  No

126

7.81 ± 3.21

 

94.3 ± 49.3

 

16.64 ± 6.80

 

CYP3A5

  *1/*1

11

10.68 ± 3.49

<0.001

64.3 ± 21.1

<0.001

16.91 ± 6.58

0.617

  *1/*3

65

9.35 ± 2.86

 

70.5 ± 24.9

 

17.18 ± 5.97

 

  *3/*3

66

5.79 ± 2.16

 

126.9 ± 54.3

 

16.06 ± 7.18

 

MDR1

  -129 T/T

130

7.83 ± 3.23

0.921

95.1 ± 49.4

0.589

16.68 ± 6.37

0.694

  -129 T/C

11

7.50 ± 3.42

 

111.0 ± 58.9

 

16.64 ± 9.19

 

  -129 C/C

1

7.00

 

84.9

 

11.00

 

MDR1

  1236 C/C

13

6.92 ± 2.52

0.218

99.4 ± 42.8

0.498

17.00 ± 6.95

0.314

  1236 C/T

77

8.21 ± 3.29

 

91.7 ± 52.1

 

17.32 ± 9.85

 

  1236 T/T

52

7.39 ± 3.22

 

102.2 ± 48.7

 

15.54 ± 7.44

 

MDR1

  2677 G/G

42

8.27 ± 3.60

0.738

94.3 ± 62.8

0.983

18.36 ± 6.77

0.206

  2677 G/T

48

7.41 ± 3.09

 

95.7 ± 43.1

 

16.04 ± 6.43

 

  2677 G/A

17

7.18 ± 2.32

 

98.4 ± 35.0

 

14.29 ± 5.97

 

  2677A/T

15

8.37 ± 3.58

 

94.7 ± 53.7

 

18.00 ± 3.40

 

  2677 T/T

19

7.87 ± 3.23

 

102.7 ± 48.1

 

15.68 ± 8.32

 

  2677A/A

1

7.00

 

69.7

 

11.00

 

MDR1

  3435 C/C

61

8.12 ± 3.21

0.568

91.4 ± 46.9

0.605

16.98 ± 6.77

0.772

  3435 C/T

61

7.51 ± 3.22

 

100.0 ± 53.7

 

16.18 ± 5.92

 

  3435 T/T

20

7.68 ± 3.32

 

99.8 ± 48.6

 

17.00 ± 8.05

 

NR1I2

  -25385 C/C

94

7.64 ± 3.25

0.335

97.8 ± 52.8

0.608

16.44 ± 6.73

0.611

  -25385 C/T

41

7.87 ± 3.18

 

95.8 ± 46.2

 

16.71 ± 6.56

 

  -25385 T/T

7

9.50 ± 2.99

 

78.2 ± 29.9

 

19.00 ± 4.76

 

Effect of MDR1 haplotypes on tacrolimus dosage requirement and the time to reach a stable dose

The MDR1 haplotypes were constructed based on the MDR1 -129T > C, MDR1 1236C > T, MDR1 2677 G > T/A and MDR1 3435C > T. Of the 142 patients, 14 halotypes and 22 diplotypes were found. We evaluated the relationships between the MDR1 diplotypes and the maintenance tacrolimus dose. Those cases of fewer than two MDR1 diplotypes were excluded from the analysis because of the insufficient statistical power. Ultimately, 134 patients were analysed using one-way ANOVA. There was no significant difference in the maintenance tacrolimus dose and the time to stable tacrolimus dosage among the MDR1 diplotypes (Table 3).
Table 3

Differences in tacrolimus dose and the time to reach a stable dose among the MDR1 diplotypes

MDR1 diplotypes

Number of patients

Stable dose (mg/day)

P value

Concentration/dose ratio (ng/ml)/(mg/kg)

P value

Time to stable dose (day)

P value

TTTT/TTTT

16

7.53 ± 3.42

 

105.9 ± 51.8

 

16.56 ± 8.77

 

TTGC/TCAC

8

8.00 ± 2.73

 

89.5 ± 37.9

 

14.75 ± 5.42

 

TTGT/TTTC

21

6.33 ± 2.89

 

109.2 ± 48.6

 

13.52 ± 7.39

 

TTGT/TCTC

16

9.03 ± 3.25

 

76.4 ± 25.7

 

18.13 ± 4.84

 

TTGC/TCGC

22

8.70 ± 3.79

 

87.2 ± 54.0

 

16.82 ± 6.34

 

TCGC/TCGC

7

7.79 ± 2.86

 

91.6 ± 45.0

 

18.14 ± 7.97

 

TCGC/TCAC

4

6.13 ± 1.93

 

108.9 ± 44.6

 

16.50 ± 7.14

 

TTTT/TCAC

7

8.07 ± 3.17

 

87.9 ± 29.6

 

18.00 ± 2.94

 

TTTT/ TTTC

3

9.67 ± 0.58

0.201

85.6 ± 13.4

0.247

11.00 ± 2.65

0.260

TTGC/TTTC

2

5.75 ± 3.18

 

148.9 ± 97.2

 

14.50 ± 0.71

 

TTTT/CCAC

5

8.60 ± 4.63

 

109.8 ± 77.8

 

16.80 ± 4.03

 

TTGT/TCGC

5

5.40 ± 2.53

 

149.7 ± 117.8

 

19.60 ± 3.65

 

TTGC/TCTC

3

6.17 ± 0.76

 

110.8 ± 8.6

 

20.67 ± 7.02

 

TTGC/TTGC

7

10.07 ± 3.38

 

66.3 ± 18.5

 

19.57 ± 4.12

 

TTGC/CCAC

3

7.83 ± 0.29

 

91.4 ± 8.4

 

8.00 ± 5.00

 

TTGT/TCTT

3

8.00 ± 4.00

 

81.1 ± 25.0

 

16.67 ± 2.08

 

TTTC/TCAC

2

11.00

 

46.5 ± 2.2

 

18.00 ± 1.41

 

Total

134

7.90 ± 3.26

 

95.5 ± 50.1

 

16.41 ± 6.38

 

Difference in concentration/dose ratio for each category variable and MDR1 haplotypes

We also analysed the effect of each category variable and MDR1 haplotype on the dose-corrected concentration. Among the eight category variables, only the CYP3A5 genotype had a significant effect on the tacrolimus concentration/dose ratio (Table 2): the tacrolimus concentration/dose ratio was higher (P < 0.001) among patients with the CYP3A5 *3/*3 genotype (126.9 ± 54.3 ng/ml/mg/kg) than in those with the CYP3A5*1/*3 (70.5 ± 24.9 ng/ml/mg/kg) or CYP3A5*1/*1 (64.3 ± 21.1 ng/ml/mg/kg) genotypes. There was no significant difference in the tacrolimus concentration/dose ratio among the MDR1 diplotypes (Table 3).

Pharmacogenetics-based tacrolimus dosing model

As shown above, of the eight category variables, only CYP3A5 genotype exhibited a significant association with daily tacrolimus dose and the dose-corrected concentration. Patients with the *3/*3 genotype exhibited a mean stable tacrolimus dose of 5.79 ± 2.16 mg/day, which was significantly lower (P < 0.001) than that of patients with the *1/*3 heterozygotic genotype (9.35 ± 2.86 mg/day) and homozygotic *1/*1 genotype (10.68 ± 3.49 mg/day). There was no significant difference (P = 0.119) in the mean stable dose between the *1/*3 heterozygotes and *1/*1 homozygotes.

The results of Pearson correlation between the continuous variables and stable tacrolimus dose (Table 4) showed that tacrolimus dose was significantly negatively correlated with haematocrit (r = −0.184, P = 0.028), haemoglobin (r = −0.197, P = 0.019), TBIL (r = −0.231,  P = 0.006), IBIL (r = −0.218, P = 0.009) and significantly positively correlated with body weight (r = 0.171, P = 0.042). There was no significant correlation between age, albumin, alanine aminotransferase (ALT), aspartate aminotransferase (AST) and tacrolimus dose.
Table 4

Univariate analysis of variables affecting tacrolimus dose requirements

Variables

r

P

Age (years)

−0.007

0.935

Body weight (kg)

0.171

0.042

Haematocrit (%)

−0.184

0.028

Haemoglobin (g/l)

−0.197

0.019

Albumin (g/l)

−0.015

0.859

ALT (U/l)

−0.116

0.171

AST (U/l)

−0.158

0.061

TBIL (μmol/l)

−0.231

0.006

IBIL (μmol/l)

−0.218

0.009

The continuous variables of age, weight and clinical factors were used as potential candidate predictors for tacrolimus dose. To select candidate predictors for the multiple regression models, we examined each continuous variable individually against tacrolimus dose requirements based on the Pearson’s correlation test. The continuous variables that were significant at P < 0.05 in the initial analyses were retained for further model building. Of the category variables, only the CYP3A5 genotype exhibited a significant effect on the tacrolimus maintenance dose. Using dummy variables to code for CYP3A5 genotype, we finally selected the body weight, haematocrit, haemoglobin, TBIL, IBIL and dummy variables for CYP3A5 genotype for the multiple regression analysis. The stepwise regression was performed to establish the tacrolimus-dosing model.

The predictors included in the final multiple linear regression model were body weight, TBIL and CYP3A5*3/*3 (R2 = 0.405; Table 5). The tacrolimus-dosing model was used to estimate maintenance daily tacrolimus doses: predicted tacrolimus dose (mg/day) = exp[1.929–0.506 ×  CYP3A5*3/*3 − 0.022 × TBIL (μmol/l) + 0.008 × body weight (kg)], with CYP3A5*3/*3 coded as 1 if present and 0 if absent.
Table 5

Multiple linear regression analysis of variables influencing tacrolimus dose requirements

Variables

Coefficient

Standard error

P value

Collinearity statistics

Tolerance

Variance inflation factor

Intercept

1.929

0.178

<0.001

CYP3A5*3/*3

−0.506

0.057

<0.001

0.984

1.017

TBIL

−0.022

0.007

0.004

0.981

1.019

Body weight

0.008

0.003

0.004

0.989

1.011

Discussion

In this study, we examined the contribution of gene polymorphisms and clinical factors to the tacrolimus maintenance dose in Chinese renal transplant recipients. Specifically, we examined the contribution of eight category variables (CYP3A5 6986A > G, MDR1 -129T > C, MDR1 1236C > T, MDR1 2677G > T/A, MDR1 3435C > T, NR1I2 -25385C > T, gender, whether or not having diabetes) and MDR1 haplotype on the dose-corrected concentration. It has been shown that the CYP3A5 6986A > G SNP can affect tacrolimus blood concentration [19, 20, 21]. Our study confirmed that the tacrolimus maintenance dose of the patients with the CYP3A5*3/*3 genotype was significantly lower than that of patients with the CYP3A5*1/*3 and CYP3A5*1/*1 genotypes. Similar results were obtained from the analysis between the CYP3A5 genotypes and the dose-corrected concentration. Consequently, in our study, the CYP3A5 6986A > G SNP significantly affected tacrolimus metabolism. Regarding the relationship between MDR1 polymorphisms and tacrolimus blood concentration, studies performed to date have obtained contradictory results. Most of these studies focused on the MDR1 1236C > T, MDR1 2677G > T/A and MDR1 3435C > T SNPs. Some of these reported that MDR1 2677 G > T/A homozygotes had lower weight-based daily tacrolimus dose requirements and higher dose-adjusted tacrolimus blood concentration than the wild-type carriers [22, 23], while other studies observed no difference for tacrolimus and MDR1 2677G > T/A [24, 25]. However, most of the studies examining the effects of the MDR1 2677G > T/A, MDR1 2677T and MDR1 2677A SNPs were usually divided in the same variant group, while one study observed that the 2677GA genotypic group had the shortest time to maximum plasma concentration after administration of a oral dose of 2 mg risperidone to patients with different genotypes of the 2677G > T/A polymorphism [26]. This phenomenon shows that the MDR1 2677T and MDR1 2677A SNPs may have different effects on MDR1 bioactivity. We therefore divided the MDR1 2677T and MDR1 2677A patients into different genotypic groups in our study and found that the P value was 0.738 for the tacrolimus dose and 0.983 for the dose-corrected concentration among the different genotypic groups. Based on this result, we concluded that there was no difference for the tacrolimus dose requirement as well as dose-corrected concentration and the MDR1 2677G > T/A. One study observed that the MDR1 -129TC genotypic group needed a relatively higher tacrolimus daily dose than the wild genotype group [22]. In our study, the MDR1 -129TC genotypic group needed a relatively lower tacrolimus daily dose than the wild genotype group and had a relatively higher tacrolimus dose-corrected concentration than the wild genotype group, although the difference was not significant. This result may suggest that more investigation on the relationship between the tacrolimus dose requirement and the MDR1 -129T > C SNP is needed. In our study, the MDR1 1236 C > T and MDR1 3435C > T SNPs had no significant effects on the tacrolimus daily dose and dose-corrected concentration, in accordance with other studies [22, 24]. Despite this, it has been reported that the haplotypes based on the MDR1 polymorphisms affect tacrolimus blood concentration [22]. Anglicheau et al. [22] reported that the tacrolimus blood concentration of renal transplant recipients with the TTT-TTT haplotype derived from the MDR1 1236C > T, MDR1 2677G > T/A and MDR1 3435C > T SNPs was significantly higher than those with the CGC-CGC haplotype. However, we did not observe this difference for the tacrolimus dose-corrected concentration and the MDR1 haplotypes, although the MDR1 -129T > C was considered in the MDR1 construction, the same as in some other studies [11, 27]. The above results may illustrate that the influence of the MDR1 haplotype on the tacrolimus dose requirement is quite limited. Benkali et al. [14] investigated the impact of the NR1I2 -25385C > T polymorphism on tacrolimus pharmacokinetics and found that the tacrolimus apparent oral clearance values were 1.2-fold and 1.5-fold higher in the NR1I2 -25385CC homozygous wildtype carriers than in CT and TT patients, respectively, suggesting that the NR1I2 -25385C > T decreases CYP3A and P-gp expression. However, our results appear to be contradictory: the tacrolimus dose requirements in the NR1I2 -25385 TT homozygous mutant-type carriers (9.50 ± 2.99 mg/day) was relatively higher than that in CT (7.87 ± 3.18 mg/day) and CC patients (7.64 ± 3.25 mg/day), although the difference is not statistically significant (P = 0.335). Similar results were obtained from the analysis of the dose-corrected concentration (P = 0.608). The association between the NR1I2-25385C > T SNP and tacrolimus metabolism needs further investigation. In this study, we also analysed the time to reach stable tacrolimus dose: there were no significant between-group differences in the time to reach stable tacrolimus dose, including between genotypic groups and haplotypic groups.

Clinical factors as well as genetic factors can affect tacrolimus metabolism. Based on the results of our univariate analysis, tacrolimus dose was significantly negatively correlated with haematocrit, haemoglobin, TBIL and IBIL and significantly positively correlated with body weight. Previous population studies found that haematocrit pharmacokinetics was related to the plasma tacrolimus concentrations [14]. Moreover, our study found a negative correlation between tacrolimus dose and haematocrit. Low haematocrit probably has a reduced proportion of tacrolimus bound to red blood cells and an increased plasma proportion, which is more readily metabolized by the liver [28]. Therefore, the low haematocrit would result in a high tacrolimus maintenance dosage. At the same time, low haematocrit values may lead to a low haemoglobin value. This possibility may be the reason why the haemoglobin was significantly negatively correlated with the tacrolimus maintenance dosage. TBIL and IBIL were significantly negatively correlated with tacrolimus dose. Previous studies also found that the TBIL was significantly positively correlated with didemethyl tacrolimus concentration [15]. Bellarosa et al. [29] reported that bilirubin is the substrate of P-gp, with bilirubin and tacrolimus competitively binding with P-gp. Consequently, the high bilirubin concentration may delay the transport and metabolism of the tacrolimus and result in a low tacrolimus dose requirement. On the other hand, a high bilirubin concentration may imply liver dysfunction, resulting in an increased tacrolimus concentration and reduced tacrolimus dose requirement.

In recent years, warfarin-dosing models based on both genetic and clinical factors have been developed to estimate the warfarin dose requirement [16, 17]. There have been no references to dosing models of cyclosporin A, the calcineurin inhibitor widely used in organ transplantation. For the estimation of tacrolimus dose requirement, a retrospective study has provided a simple method for individualizing the first oral dose of tacrolimus on the basis of CYP3A5 genotype. The authors proposed that 0.075 mg/kg twice a day should be given to CYP3A5 nonexpressors, while a double dose of 0.150 mg/kg twice a day should be administered to CYP3A5 expressors [7]. A subsequent large prospective trial demonstrated that this simple method for tacrolimus maintenance dose prediction could shorten the time to achieve the target blood concentration post-transplantation [30]. Other mathematical models are focused on the predication of tacrolimus blood concentration but not dosage [14, 18, 31]. However, our tacrolimus-dosing model not only considers genetic factors, but also takes clinical factors into account. The patients with the same CYP3A5 genotype may have the different predicted tacrolimus maintenance dose; therefore our dosing model is more reliable and significant for use in clinical practice.

We excluded 61 patients from participating in this study due to disease states which may affect tacrolimus pharmacokinetics and pharmacodynamics. The exclusion of the patients with some disease states is necessary because those diseases may affect tacrolimus metabolism and the results of the study. Patients with hepatitis B or hepatitis C may have liver dysfunction and therefore need a low tacrolimus maintenance dose. Patients who have received second renal transplantation and those who have had cancer may have a different internal environment from other patients. We were uncertain about the effects of donor livers on those patients who had received both liver and renal transplantation so they were also excluded. Hormones have been reported to be capable of inducing the expression of CYP3A and MDR1 genes [13]; therefore, SLE patients with a long-term hormone therapy should be excluded owing to the uncertainty of the induced condition of CYP3A and MDR1 genes. In addition, the aims of the tacrolimus-dosing model were to estimate the maintenance tacrolimus dose and reduce the time in which the tacrolimus trough blood concentration can reach the predefined therapeutic range. Hence, only those patients who had reached the stable maintenance tacrolimus dose were included for the development of the dosing model, resulting in the exclusion of 24 patients whose target trough blood concentration did not reach the predefined therapeutic range. The exclusion of these patients may have two implications: (1) the strict selection of patients made the results more reliable and the dosing model more delicate; (2) the question of whether or not the dosing model can be used in patients with the excluded disease states needs more investigation. Meanwhile, a small percentage of patients without these diseases may hardly reach the predefined tacrolimus therapeutic range even using the tacrolimus-dosing model.

In conclusion, our study has examined the impacts of the genetic factors and clinical factors on the stable tacrolimus dose requirement and the time to reach stable maintenance dose. We also analysed the impacts of eight category variables and MDR1 haplotype on the dose-corrected concentration. Finally, we developed a tacrolimus-dosing model that estimated the dose requirement for patients according to these factors. However, this model should be validated in large prospective trials in multiple centres across China, if possible. We hope that this dosing model can be improved and become useful to clinicians in their aim to use tacrolimus more safely and effectively.

Notes

Acknowledgments

This work was supported by the Research Foundation for the President of Nanfang Hospital to Chuan-Jiang Li.

References

  1. 1.
    Ekberg H, Tedesco-Silva H, Demirbas A, Vítko S, Nashan B, Gürkan A, Margreiter R, Hugo C, Grinyó JM, Frei U, Vanrenterghem Y, Daloze P, Halloran PF, ELITE-Symphony Study (2007) Reduced exposure to calcineurin inhibitors in renal transplantation. N Engl J Med 357:2562–75PubMedCrossRefGoogle Scholar
  2. 2.
    Kuypers DRJ, Claes K, Evenepoel P, Maes B, Vanrenterghem Y (2004) Clinical efficacy and toxicity profile of tacrolimus and mycophenolic acid in relation to combined long-term pharmacokinetics in de novo renal allograft recipients. Clin Pharmacol Ther 75:434–47PubMedCrossRefGoogle Scholar
  3. 3.
    Masuda S, Inui K (2006) An up-date review on individualized dosage adjustment of calcineurin inhibitors in organ transplant patients. Pharmacol Ther 112:184PubMedCrossRefGoogle Scholar
  4. 4.
    Wallemacq P, Armstrong VW, Brunet M, Haufroid V, Holt DW, Johnston A, Kuypers D, Le Meur Y, Marquet P, Oellerich M, Thervet E, Toenshoff B, Undre N, Weber LT, Westley IS, Mourad M (2009) Opportunities to optimize tacrolimus therapy in solid organ transplantation: report of the European consensus conference. Ther Drug Monit 31:139–52PubMedCrossRefGoogle Scholar
  5. 5.
    Xie HG, Wood AJ, Kim RB, Stein CM, Wilkinson GR (2004) Genetic variability in CYP3A5 and its possible consequences. Pharmacogenomics 5:243–72PubMedCrossRefGoogle Scholar
  6. 6.
    Roy JN, Barama A, Poirier C, Vinet B, Roger M (2006) CYP3A4, CYP3A5, and MDR-1 genetic influences on tacrolimus pharmacokinetics in renal transplant recipients. Pharmacogenet Genomics 16:659–65PubMedCrossRefGoogle Scholar
  7. 7.
    Haufroid V, Wallemacq P, VanKerckhove V, Elens L, De Meyer M, Eddour DC, Malaise J, Lison D, Mourad M (2006) CYP3A5 and ABCB1 polymorphisms and tacrolimus pharmacokinetics in renal transplant candidates: guidelines from an experimental study. Am J Transplant 6:2706–13PubMedCrossRefGoogle Scholar
  8. 8.
    Ambudkar SV, Kimchi-Sarfaty C, Sauna ZE, Gottesman MM (2003) P-glycoprotein: from genomics to mechanism. Oncogene 22:7468–85PubMedCrossRefGoogle Scholar
  9. 9.
    Hesselink DA, van Gelder T, van Schaik RH (2005) The pharmacogenetics of calcineurin inhibitors: one step closer toward individualized immunosuppression? Pharmacogenomics 6:323–37PubMedCrossRefGoogle Scholar
  10. 10.
    Wang J, Zeevi A, McCurry K, Schuetz E, Zheng H, Iacono A, McDade K, Zaldonis D, Webber S, Watanabe RM, Burckart GJ (2006) Impact of ABCB1 (MDR1) haplotypes on tacrolimus dosing in adult lung transplant patients who are CYP3A5 *3/*3 non-expressors. Transpl Immunol 15:235–40PubMedCrossRefGoogle Scholar
  11. 11.
    Mai I, Perloff ES, Bauer S, Goldammer M, Johne A, Filler G, Budde K, Roots I (2004) MDR1 haplotypes derived from exons 21 and 26 do not affect the steady-state pharmacokinetics of tacrolimus in renal transplant patients. Br J Clin Pharmacol 58:548–53PubMedCrossRefGoogle Scholar
  12. 12.
    Fredericks S, Moreton M, Reboux S, Carter ND, Goldberg L, Holt DW, MacPhee IA (2006) Multidrug resistance gene-1 (MDR-1) haplotypes have a minor influence on tacrolimus dose requirements. Transplantation 82:705–8PubMedCrossRefGoogle Scholar
  13. 13.
    Rosenfeld JM, Vargas R Jr, Xie W, Evans RM (2003) Genetic profiling defines the xenobiotic gene network controlled by the nuclear receptor pregnane X receptor. Mol Endocrinol 17:1268–82PubMedCrossRefGoogle Scholar
  14. 14.
    Benkali K, Prémaud A, Picard N, Rérolle JP, Toupance O, Hoizey G, Turcant A, Villemain F, Le Meur Y, Marquet P, Rousseau A (2009) Tacrolimus population pharmacokinetic-pharmacogenetic analysis and Bayesian estimation in renal transplant recipients. Clin Pharmacokinet 48:805–16PubMedCrossRefGoogle Scholar
  15. 15.
    Gonschior AK, Christians U, Winkler M, Linck A, Baumann J, Sewing KF (1996) Tacrolimus (FK506) metabolite patterns in blood from liver and kidney transplant patients. Clin Chem 42:1426–32PubMedGoogle Scholar
  16. 16.
    Sconce EA, Khan TI, Wynne HA, Avery P, Monkhouse L, King BP, Wood P, Kesteven P, Daly AK, Kamali F (2005) The impact of CYP2C9 and VKORC1 genetic polymorphism and patient characteristics upon warfarin dose requirements: proposal for a new dosing regimen. Blood 106:2329–33PubMedCrossRefGoogle Scholar
  17. 17.
    Tham LS, Goh BC, Nafziger A, Guo JY, Wang LZ, Soong R, Lee SC (2006) A warfarin dosing model in Asians that uses single-nucleotide polymorphisms in vitamin K epoxide reductase complex and cytochrome P450 2C9. Clin Pharmacol Ther 80:346–55PubMedCrossRefGoogle Scholar
  18. 18.
    Jin Z, Zhang WX, Chen B, Mao AW, Cai WM (2009) Stepwise regression analysis of the determinants of blood tacrolimus concentrations in Chinese patients with liver transplant. Med Chem 5:301–4PubMedCrossRefGoogle Scholar
  19. 19.
    Thervet E, Anglicheau D, King B, Schlageter MH, Cassinat B, Beaune P, Legendre C, Daly AK (2003) Impact of cytochrome p450 3A5 genetic polymorphism on tacrolimus doses and concentration-to-dose ratio in renal transplant recipients. Transplantation 76:1233–5PubMedCrossRefGoogle Scholar
  20. 20.
    Katsakiori PF, Papapetrou EP, Sakellaropoulos GC, Goumenos DS, Nikiforidis GC, Flordellis CS (2010) Factors affecting the long-term response to tacrolimus in renal transplant patients: pharmacokinetic and pharmacogenetic approach. Int J Med Sci 7:94–100PubMedGoogle Scholar
  21. 21.
    Suzuki Y, Homma M, Doki K, Itagaki F, Kohda Y (2008) Impact of CYP3A5 genetic polymorphism on pharmacokinetics of tacrolimus in healthy Japanese subjects. Br J Clin Pharmacol 66:154–5PubMedCrossRefGoogle Scholar
  22. 22.
    Anglicheau D, Verstuyft C, Laurent-Puig P, Becquemont L, Schlageter MH, Cassinat B, Beaune P, Legendre C, Thervet E (2003) Association of the multidrug resistance-1 gene single-nucleotide polymorphisms with the tacrolimus dose requirements in renal transplant recipients. J Am Soc Nephrol 14:1889–96PubMedCrossRefGoogle Scholar
  23. 23.
    Mendes J, Martinho A, Simoes O, Mota A, Breitenfeld L, Pais L (2009) Genetic polymorphisms in CYP3A5 and MDR1 genes and their correlations with plasma levels of tacrolimus and cyclosporine in renal transplant recipients. Transplant Proc 41:840–2PubMedCrossRefGoogle Scholar
  24. 24.
    Tsuchiya N, Satoh S, Tada H, Li Z, Ohyama C, Sato K, Suzuki T, Habuchi T, Kato T (2004) Influence of CYP3A5 and MDR1 (ABCB1) polymorphisms on the pharmacokinetics of tacrolimus in renal transplant recipients. Transplantation 78:1182–7PubMedCrossRefGoogle Scholar
  25. 25.
    Haufroid V, Wallemacq P, VanKerckhove V, Elens L, De Meyer M, Eddour DC, Malaise J, Lison D, Mourad M (2006) CYP3A5 and ABCB1 polymorphisms and tacrolimus pharmacokinetics in renal transplant candidates: guidelines from an experimental study. Am J Transplant 6:2706–13PubMedCrossRefGoogle Scholar
  26. 26.
    Xiang Q, Zhao X, Zhou Y, Duan JL, Cui YM (2010) Effect of CYP2D6, CYP3A5, and MDR1 genetic polymorphisms on the pharmacokinetics of risperidone and its active moiety. J Clin Pharmacol 50:659–66PubMedCrossRefGoogle Scholar
  27. 27.
    Choi JH, Lee YJ, Jang SB, Lee JE, Kim KH, Park K (2007) Influence of the CYP3A5 and MDR1 genetic polymorphisms on the pharmacokinetics of tacrolimus in healthy Korean subjects. Br J Clin Pharmacol 64:185–91PubMedCrossRefGoogle Scholar
  28. 28.
    Undre NA, Schäfer A (1998) Factors affecting the pharmacokinetics of tacrolimus in the first year after renal transplantation. European Tacrolimus Multicentre Renal Study Group. Transplant Proc 30:1261–3PubMedCrossRefGoogle Scholar
  29. 29.
    Bellarosa C, Bortolussi G, Tiribelli C (2009) The role of ABC transporters in protecting cells from bilirubin toxicity. Curr Pharm Des 15:2884–92PubMedCrossRefGoogle Scholar
  30. 30.
    Thervet E, Loriot MA, Barbier S, Buchler M, Ficheux M, Choukroun G, Toupance O, Touchard G, Alberti C, Le Pogamp P, Moulin B, Le Meur Y, Heng AE, Subra JF, Beaune P, Legendre C (2010) Optimization of initial tacrolimus dose using pharmacogenetic testing. Clin Pharmacol Ther 87:721–6PubMedGoogle Scholar
  31. 31.
    Mathew BS, Fleming DH, Jeyaseelan V, Chandy SJ, Annapandian VM, Subbanna PK, John GT (2008) A limited sampling strategy for tacrolimus in renal transplant patients. Br J Clin Pharmacol 66:467–72PubMedCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  • Liang Li
    • 1
  • Chuan-Jiang Li
    • 2
  • Lei Zheng
    • 4
  • Yan-Jun Zhang
    • 3
  • Hai-Xia Jiang
    • 4
  • Bo Si-Tu
    • 4
  • Zhong-Hai Li
    • 2
  1. 1.Department of Medical Genetics, School of Basic Medical SciencesSouthern Medical UniversityGuangzhouPeoples Republic of China
  2. 2.Department of Organ Transplantation, Nanfang HospitalSouthern Medical UniversityGuangzhouPeoples Republic of China
  3. 3.College of PharmacyUniversity of Cincinnati Academic Health CentreCincinnatiUSA
  4. 4.Department of clinical laboratory, Nanfang HospitalSouthern Medical UniversityGuangzhouPeoples Republic of China

Personalised recommendations