Effects of the CYP3A4*1B Genetic Polymorphism on the Pharmacokinetics of Tacrolimus in Adult Renal Transplant Recipients: A Meta-Analysis

Background and Objective The association between the CYP3A4*1B single nucleotide polymorphism (SNP) and tacrolimus pharmacokinetics in different studies is controversial. Therefore, a meta-analysis was employed to evaluate the correlation between the CYP3A4*1B genetic polymorphism and tacrolimus pharmacokinetics at different post-transplantation times in adult renal transplant recipients. Methods Studies evaluating the CYP3A4*1B genetic polymorphism and tacrolimus pharmacokinetics were retrieved through a systematical search of Embase, PubMed, the Cochrane Library, ClinicalTrials.gov and three Chinese literature databases (up to Sept. 2014). The pharmacokinetic parameters (weight-adjusted tacrolimus daily dose and tacrolimus trough concentration/weight-adjusted tacrolimus daily dose ratio) were extracted, and the meta-analysis was performed using Stata 12.1. Results Seven studies (involving 1182 adult renal transplant recipients) were included in this meta-analysis. For the weight-adjusted tacrolimus daily dose, in all included renal transplant recipients (European & Indian populations), CYP3A4*1/*1 recipients required a significantly lower weight-adjusted tacrolimus daily dose than did CYP3A4*1B carriers at 7 days (WMD -0.048; 95% CI -0.083 ~ -0.014), 6 months (WMD -0.058; 95% CI -0.081 ~ -0.036) and 12 months (WMD - 0.061; 95% CI -0.096 ~ -0.027) post-transplantation. In light of the heterogeneity, the analysis was repeated after removing the only study in an Indian population, and CYP3A4*1/*1 European recipients (mostly Caucasian) required a lower weight-adjusted tacrolimus daily dose within the first year post-transplantation. The tacrolimus trough concentration/weight-adjusted tacrolimus daily dose ratio (C0/Dose ratio) was significantly higher in CYP3A4*1/*1 recipients than in CYP3A4*1B carriers at 6 months (WMD 52.588; 95% CI 22.387 ~ 82.789) and 12 months (WMD 62.219; 95% CI 14.218 ~ 110.221) post-transplantation. When the only study in an Indian population was removed to examine European recipients (mostly Caucasian), the significant difference persisted at 1 month, 6 months and 12 months post-transplantation. Conclusion Based on our meta-analysis, the CYP3A4*1B genetic polymorphism affects tacrolimus dose requirements and tacrolimus trough concentration/weight-adjusted tacrolimus daily dose ratio within the first year post-transplantation in adult renal transplant recipients, especially in European recipients (mostly Caucasian).


Introduction
Renal transplantation is an effective treatment for the patients with end-stage renal disease. Tacrolimus, a macrolide antibiotic compound, is the most frequently used maintenance immunosuppressant after renal transplantation [1]. However, tacrolimus is characterized by its narrow therapeutic index and significant inter-individual variability in pharmacokinetics. Tacrolimus blood concentration below target trough levels can lead to rejection, and higher trough blood concentrations can lead to toxicity and infection [2,3]. Achieving a steady target blood concentration is critical to avoid rejection and adverse drug effects [4]. However, several factors influence the pharmacokinetics of tacrolimus, including hepatic dysfunction, posttransplantation time, hematocrit, serum albumin, age, race and drug interactions, especially gene polymorphism [5]. Single nucleotide polymorphisms (SNPs) in cytochrome P450 3A (CYP3A) play an important role in tacrolimus metabolism [6]. CYP3A enzymes in human liver microsomes play a major role in the oxidation of tacrolimus [7], and the tacrolimus metabolism within the small intestinal contributes significantly to its bioavailability [8,9]. Many studies in renal transplant recipients focus on CYP3A5 Ã 3 genetic polymorphism (rs776746, 6986A>G). There is a widespread view that CYP3A5 nonexpressers (CYP3A5 Ã 3/ Ã 3 carriers) required lower mean tacrolimus doses [10] and exhibit higher trough concentration/dose ratios [11,12]. The CYP3A4 Ã 1B genetic polymorphism (rs2740574, −392A>G), linked to enhanced CYP3A4 activity, is likely related to the rapid metabolism of tacrolimus [6], but the effect of the CYP3A4 Ã 1B genetic polymorphism on tacrolimus pharmacokinetics (dose and concentration) in renal transplant recipients is controversial [13], and there has been no meta-analysis to assess the issue to date.
To evaluate the correlation between the CYP3A4 Ã 1B genetic polymorphism and tacrolimus pharmacokinetics (weight-adjusted tacrolimus daily dose and tacrolimus trough concentration/weight-adjusted tacrolimus daily dose ratio), a meta-analysis was employed to systematically review the published evidence of the relationship between the CYP3A4 Ã 1B genetic polymorphism and tacrolimus pharmacokinetics in adult renal transplant recipients.
2014, without language and publication status restrictions, for published studies that evaluated the effects of the CYP3A4 Ã 1B genetic polymorphism on tacrolimus pharmacokinetics. The search terms (("tacrolimus" or "FK506") and "CYP3A4") as well as related Chinese keywords in the Chinese databases were used. In addition, the reference lists of the included articles and relevant reviews were searched manually. In cases of missing data, the original authors were contacted for more detailed information by e-mail.
The inclusion criteria for the included studies were as follows: (a) studies focus on the effects of the CYP3A4 Ã 1B genetic polymorphism on adult renal transplant recipients treated with tacrolimus; (b) tacrolimus pharmacokinetics parameters was described separately according to different CYP3A4 Ã 1B genotypes; and (c) tacrolimus pharmacokinetic parameters were measured at explicit post-transplantation times. According to the above criteria, studies were assessed independently by two reviewers (S.W.L. and T.H.L.) for inclusion in the metaanalysis.

Data extraction and quality assessment
Relevant data from all eligible studies were extracted independently by the two reviewers (S.W. L. and T.H.L.), and discrepancies in the data extraction were resolved through consensus. The following information was collected: first author, publication information, design of the study, demographic data, immunosuppressive protocol, method of concentration measured, genotype frequency, post-transplantation time, weight-adjusted tacrolimus daily dose (Dose), tacrolimus trough concentration (C 0 ), C 0 /Dose ratio. For continuous data, information was collected as mean (SD), if the studies provided the median (range), the method reported by Hozo et al. [14] was employed to estimate the mean (SD).
The quality of the included studies was assessed by two reviewers (S.W.L. and T.H.L.) through a checklist derived from the Strengthening the Reporting of Genetic Association (STREGA) recommendations for reports on genetic association studies [15], and modified according to the quality checklist described elsewhere [16,17].

Statistical analysis
The Dose and C 0 /Dose ratio values were compared between CYP3A4 Ã 1/ Ã 1 recipients and CYP3A4 Ã 1B carriers, and a random-effect model was used for all meta-analyses. The data of the CYP3A4 Ã 1B carriers were calculated from the CYP3A4 Ã 1/ Ã 1B and the CYP3A4 Ã 1B/ Ã 1B groups using the method provided by Table 7.7.a of the Cochrane handbook 5.1.0 [18]. The continuous data were pooled by weighted mean difference (WMD) or standard mean difference (SMD) and 95% confidence interval (CI), and Z-tests were performed to determine the statistical significance of the results. Statistical significance was defined as P < 0.05.
The heterogeneity across the included studies was assessed using the I 2 statistic, with significance defined as I 2 > 50%. In case of substantial heterogeneity (I 2 > 50%), meta-regression was performed to explore the sources of heterogeneity [post-transplantation time (7 days, 1 month, 3 months, 6 months, 12 months), ethnicity (Caucasian, Indian, mixed race), location (Europe, India), method of concentration measured (MEIA, CMIA, EMIT), initial dose (0.1-0.16 mg/kg/day, 0.2-0.3 mg/kg/day), and Hardy-Weinberg equilibrium (equilibrium or disequilibrium)]. Further subgroup analysis was performed according to the results of the metaregression. A sensitivity analysis was performed to assess the validity of the outcomes by excluding each observation successively. A publication bias analysis was not performed because less than 10 studies were included. All statistical analysis was performed using Stata 12.1.

Characteristics of the articles included in the meta-analysis
A total of 683 publications were identified by the literature search. After screening the titles and abstracts, the full texts of the remaining 93 studies were further assessed, and 7 studies [19][20][21][22][23][24][25] were included in the final meta-analysis. The details of identification of the eligible studies and the reasons for the exclusion of studies are presented in Fig 1. All 7 studies were published in English. A total of 1182 adult renal transplant recipients were included in the studies, and the characteristics of the 7 included studies are presented in Table 1. The results of quality assessment are presented in Table 2. Five original authors [20,21,[23][24][25] were contacted for missing or specific demographic data, and three authors [20,21,25] replied.
Effects of the CYP3A4*1B genetic polymorphism on the weight-adjusted tacrolimus dose (Dose) All 7 studies [19][20][21][22][23][24][25] evaluated the association between the CYP3A4 Ã 1B genetic polymorphism and weight-adjusted tacrolimus daily dose (Dose) at different post-transplantation time. The result of the meta-analysis revealed that CYP3A4 Ã 1/ Ã 1 recipients required a lower Dose than CYP3A4 Ã 1B carriers (WMD -0.047; 95% CI -0.062~-0.031; P < 0.001). However, substantial heterogeneity existed (I 2 = 75.3%), and a meta-regression was performed to explore the sources of heterogeneity with respect to the following factors: post-transplantation time, ethnicity, location, method of concentration measured, initial dose and Hardy-Weinberg equilibrium. The results of the meta-regression are presented in Table 3. The tacrolimus daily dose varied by post-transplantation time, ethnicity and location, although only "method of concentration measured" (r = 0.027, P = 0.008) and "initial dose" (r = -0.077, P < 0.001) contributed to heterogeneity. We performed subgroup analyses for all of the above covariates with the exception of the "Hardy-Weinberg equilibrium" because of collinearity, and the results are presented in Table 4. As Table 4 indicated, although the heterogeneity persisted, most subgroups demonstrated that CYP3A4 Ã 1/ Ã 1 recipients required a lower Dose than CYP3A4 Ã 1B carriers, but there was no statistical significance detected in subgroup analyses by post-transplantation time (1 month & 3 months), ethnicity (North Indian), location (India) or initial dose(0.1-0.16 mg/kg/ day) involving the India population.
Effects of the CYP3A4*1B genetic polymorphisms on Dose and the C 0 / Dose ratio stratified by the CYP3A5 genotype We performed subgroup analyses at different post-transplantation times stratified by the CYP3A5 genotype. In CYP3A5 Ã 3/ Ã 3 recipients, CYP3A4 Ã 1/ Ã 1 recipients required lower Dose than CYP3A4 Ã 1B carriers at 7 days, 6 months and 12 months post-transplantation; in CYP3A5 Ã 1 carriers, CYP3A4 Ã 1/ Ã 1 recipients required lower Dose than CYP3A4 Ã 1B carriers at 6 months and 12 months post-transplantation. Except for CYP3A5 Ã 3/ Ã 3 recipients at the time point of 6 months post-transplantation, there was no significant difference in the C 0 /Dose ratio between the CYP3A4 Ã 1/ Ã 1 recipients and CYP3A4 Ã 1B carriers between different CYP3A5 genotypes ( Table 5). The results suggested that the effect of the CYP3A4 Ã 1B genetic polymorphism on tacrolimus pharmacokinetics was independent at 7 days, 6 months and 12 months posttransplantation in CYP3A5 Ã 3/ Ã 3 recipients; and at 6 months and 12 months post-transplantation in CYP3A5 Ã 1 carriers.

Discussion
The findings of our meta-analysis suggest that the CYP3A4 Ã 1B genetic polymorphism influences the weight-adjusted tacrolimus daily dose and the C 0 /Dose ratio in adult renal transplant recipients. In all included renal transplant recipients, relative to CYP3A4 Ã 1B carriers, CYP3A4 Ã 1/ Ã 1 recipients required a lower weight-adjusted tacrolimus daily dose (Dose) at 7 days, 6 months and 12 months post-transplantation, and exhibited a higher C 0 /Dose ratio at 6 and 12 months post-transplantation. This finding suggests that CYP3A4 Ã 1/ Ã 1 recipients required a lower initial dose within 7 days post-transplantation and a lower maintenance dose to achieve the target blood concentration relative to CYP3A4 Ã 1B carriers at 6 and 12 months post-transplantation.
In the meta-regression and subgroup analyses, we determined that the ethnicity and location influenced the pooled estimate results significantly. Therefore, the meta-analysis was stratified by post-transplantation time and was performed in the European recipients. The results revealed that CYP3A4 Ã 1/ Ã 1 recipients exhibited a lower weight-adjusted tacrolimus daily dose (Dose) within the entire first year and a higher C 0 /Dose ratio at 1 month, 6 months and 12 months post-transplantation compared with CYP3A4 Ã 1B carriers. Thus, the CYP3A4 Ã 1B genetic polymorphism plays a more important role in European renal transplant recipients. The results of the sensitivity analyses were consistent with the meta-regression results; excluding the Indian population changed the pooled estimate at 1 month and 3 months with respect to Dose and at 1 month with respect to the C 0 /Dose ratio post-transplantation, but there was no effect on the overall estimate. Although there was no significant difference in the C 0 /Dose ratio between CYP3A4 Ã 1/ Ã 1 recipients and CYP3A4 Ã 1B carriers stratified by CYP3A5 genotype, the dose requirement differed significantly at 6 and 12 months post-transplantation, which indicated that different doses were required in CYP3A4 Ã 1/ Ã 1 recipients and CYP3A4 Ã 1B carriers stratified by the CYP3A5 genotype; these data further indicate that CYP3A4 Ã 1B and CYP3A5 Ã 3 may have independent effects on tacrolimus pharmacokinetics.
In clinical settings, the initial tacrolimus dose is given according to the weight of different renal recipients, and the maintenance dose is adjusted according to the blood concentration     (1) [26]. Acute rejection and nephrotoxicity are unavoidable, especially in the early post-transplantation stage because the clinicians must adjust the dose frequently to achieve the target blood concentration. According to our meta-analysis, the CYP3A4 Ã 1B genetic polymorphism should be considered when determining the initial tacrolimus dose and adjusting the maintenance dose, which may be helpful to achieve the target concentration in a shorter time and reduce the concentration fluctuations. Because of the therapeutic drug monitoring (TDM), even though the CYP3A4 Ã 1B carriers required a higher Dose than the CYP3A4 Ã 1/ Ã 1 recipients, the C 0 /Dose ratio were not stable within 3 months post-transplantation. Several limitations should be noted in our meta-analysis. First, only 7 observational studies (involving 1182 adult renal transplant recipients) were included, and there were only 77 CYP3A4 Ã 1B carriers. Considering the influence of large sample size from one single study, a sensitivity analysis had been performed to assess the validity of the pooled estimates. There was no single study which have a significant influence on the pooled estimate except for the study in the Indian population, which had been analyzed in Results. And our meta-analysis had tried to include all the studies that evaluated the effects of the CYP3A4 Ã 1B genetic polymorphism on tacrolimus pharmacokinetics, which could be retrieved from electronic database at present. The limitation of sample size influenced the pooled estimate possibly. Because of only 7 studies were included, the publication bias analysis was not performed. Second, the formulae for estimating the mean using the values of the median provided by Hozo et al. introduced some uncertainty [14], and the formulae for combining groups provided by the Cochrane handbook 5.1.0 Table 7.7.a provided a slight underestimate of the desired standard deviation [18]; bias resulting from the methods is unavoidable. Third, the results of our meta-analysis should be interpreted with caution because of the substantial heterogeneity across all current available studies, even though we performed meta-regression to explore the source of heterogeneity and subgroup analyses to minimize the heterogeneity. According to our meta-regression results, the method of concentration measured was the source of heterogeneity, because different methods of concentration measured have different properties in the cross-reaction with tacrolimus metabolites [27], a further subgroup analyses by method of concentration measured revealed that CYP3A4 Ã 1B carriers required higher Dose in all three subgroups, and had lower C 0 / Dose in EMIT subgroup (Table 4). Although the included studies had stated the demographic information and immunosuppressive protocol, different steroid tapering schedules may have influenced on the pooled estimate due to the interaction between steroids and tacrolimus [28]. Therefore, we listed the SMD of the pooled estimate to minimize the influence of the method of concentration measured and the steroids tapering schedules in various studies, which should be considered when the conclusions were interpreted by different transplantation centers based on different situations. Futhermore, even though several studies [13] had demonstrated linkage disequilibrium between CYP3A4 Ã 1B and CYP3A5 Ã 1, in the combined CYP3A4 Ã 1B/CYP3A5 Ã 1 genotype analysis, Gervasini, G et al. [21] and Chitnis, S D et al. [29] reported that CYP3A4 Ã 1/ Ã 1 recipients exhibited higher tacrolimus trough concentrations and C 0 /Dose ratios than CYP3A4 Ã 1B carriers in CYP3A5 expressers (CYP3A5 Ã 1/ Ã 1 or CYP3A5 Ã 1/ Ã 3); Tavira, B et al. [22] reported that, relative to CYP3A4 Ã 1B carriers stratified by CYP3A5 (expressers or non-expressers respectively), CYP3A4 Ã 1/ Ã 1 recipients required lower tacrolimus doses, and exhibited higher tacrolimus trough concentrations and C 0 /dose ratios, which suggests a significant role of the CYP3A4 Ã 1B. Our meta-analysis revealed the CYP3A4 Ã 1B and CYP3A5 Ã 3 may have independent effects on tacrolimus pharmacokinetics. Furthermore, tacrolimus is a substrate for both P-glycoprotein (P-gp, coded by ABCB1) and CYP3A, ABCB1 genetic polymorphisms (such as C3435T) have an influence on the activity of P-gp and tacrolimus pharmacokinetics [30]. The intestinal CYP3A4 and P-gp work together in a coordinated manner to serve as an absorption barrier against tacrolimus [31], and the altered activity of P-gp has a significant influence on tacrolimus metabolism by CYP3A4 in both gut and liver [32,33]. However, lack of studies focused on the interaction between ABCB1 and CYP3A4 genetic polymorphisms on tacrolimus in reanl transplant recipients, which limited analyzing the issues further. It is possible that a combination of CYP3A4/5 [34] and ABCB1 genotypes would be more helpful in making predictions than any single gene.
In conclusion, our meta-analysis suggests that the CYP3A4 Ã 1B genetic polymorphism may affect the tacrolimus dose requirements and the C 0 /Dose ratio within the first year posttransplantation in adult renal transplant recipients, especially in European recipients (mostly Caucasian). CYP3A4 genotyping before transplantation would be helpful to provide an appropriate initial dose and adjust the maintenance dose in adult renal transplant recipients.