Association of P450 Oxidoreductase Gene Polymorphism with Tacrolimus Pharmacokinetics in Renal Transplant Recipients: A Systematic Review and Meta-Analysis

There are conflicting results regarding the effect of the P450 oxidoreductase (POR) *28 genotype on the tacrolimus (TAC) pharmacokinetics (PKs) during the early post-transplantation period in adult renal transplant recipients. Thus, we characterized the impact of POR*28 on TAC PKs. We conducted a systematic review on the association between POR*28 and PKs of TAC in adult renal transplant recipients. Structured searches were conducted using PubMed, Web of Science, and Embase. TAC standardized trough concentration (ng/mL per mg/kg) data were extracted. Mean differences (MD) and their corresponding 95% confidence intervals (CIs) were used to identify the differences between the POR*28 genotype and PKs of TAC. The subgroup analysis was conducted according to CYP3A5 expression status. Six studies (n = 1061) were included. TAC standardized trough concentrations were significantly lower in recipients with the POR*28 allele compared to recipients with POR*1/*1 (MD: 8.30 ng/mL per mg/kg; 95% CI: 1.93, 14.67; p = 0.01). In the subgroup analysis, TAC standardized trough concentrations were lower for subjects who were POR*28 carriers than those who were POR*1/*1 in CYP3A5 expressers (MD: 20.21 ng/mL per mg/kg; 95% CI: 16.85, 23.56; p < 0.00001). No significant difference between POR*28 carriers and POR*1/*1 was found in the CYP3A5 non-expressers. The results of our meta-analysis demonstrated a definite correlation between the POR*28 genotype and PKs of TAC. Patients carrying the POR*28 allele may require a higher dose of TAC to achieve target levels compared to those with POR*1/*1, especially in CYP3A5 expressers.


Introduction
Tacrolimus (TAC), one of the calcineurin inhibitors, is commonly used as an immunosuppressant to prevent acute organ rejection after kidney transplantation [1]. TAC has a narrow therapeutic index and wide interindividual pharmacokinetic (PK) variability. Thus, TAC administration requires therapeutic drug monitoring (TDM) to enhance efficacy and to avoid side effects [1][2][3][4]. Although TDM is widely practiced, some patients are exposed to sub-or supra-therapeutic concentrations of TAC, thereby increasing their risk of acute organ rejection or toxicity within a week after transplantation [5].
Recently, further attention has been given to P450 oxidoreductase (POR), which transfers electrons from nicotinamide-adenine-dinucleotide phosphate-oxidase to CYP enzymes, inducing CYP expression and affecting TAC metabolism [13,14]. Among several single nucleotide polymorphisms (SNPs) of POR, the most common variant is POR*28 (c.1508 C>T, rs1057868). According to an in vitro study, this SNP was associated with increased CYP activity, including CYP1A2, CYP2C19, CYP3A4, and CYP3A5 [15]. Previous studies have investigated the role of POR*28 in the PKs of TAC and reported that patients carrying POR*28 exhibited lower trough concentrations of TAC and required higher TAC doses than wild-type patients (POR*1/*1) [16][17][18]. However, the results of previous studies are conflicting due to their small sample sizes. Therefore, we conducted a systematic review and meta-analysis of the existing studies to determine the effects of POR*28 on TAC trough concentrations in renal transplant patients.

Search Strategy and Study Selection
This study was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [19]. We performed a comprehensive search of three electronic databases (PubMed, Web of Science, and Embase) on 16 July 2021 using the following search terms: (tacrolimus OR FK506 OR FK-506 OR (calcineurin inhibitor) OR Prograf OR immunosuppress*) AND ((kidney transplant*) OR (kidney graft*) OR (kidney allograft*) OR (renal transplant*) OR (renal graft*) OR (renal allograft*)) AND (POR OR (p450 oxidoreductase) OR (cytochrome p450 oxidoreductase) OR CYPOR) AND (polymorph* OR variant* OR mutation* OR genotyp* OR phenotyp* OR haplotyp* OR SNP OR rs1057868 OR Ala503Val OR A503V) ( Table 1). Studies were selected if (1) the studies focused on the effects of the POR*28 genotype on renal transplant patients receiving TAC; (2) the studies had TAC PK data expressed as standardized trough concentration (ng/mL per mg/kg); and (3) the articles were published in English. Standardized trough concentration was determined as the concentration adjusted by the dose per body weight. Studies were excluded if they were (1) abstracts, reviews, editorials, or letters; (2) in vitro or in vivo studies; (3) studies performed on pediatric patients; or (4) studies from which we were unable to extract outcome data.
After removing duplicate studies, two authors independently excluded irrelevant studies by reviewing the titles and abstracts. Then, full-text articles were assessed for inclusion. Any inconsistencies were resolved by consensus between the two authors.

Data Extraction and Study Quality Assessment
Two reviewers independently extracted data using a preconceived data extraction spreadsheet. The following information was included: name of the first author, publication year, ethnicity, patient numbers, percentage of males, mean age, mean body weight, followup day, TAC initial dose, target trough level, concomitant drugs, and method of genotyping and quantification. Two reviewers assessed the study's quality using the Newcastle-Ottawa scale (NOS) tool [20]. The NOS tool is based on three domains: the selection of exposed and unexposed subjects (0-4 points), comparability of study groups (0-2 points), and outcome assessment (0-3 points). In terms of comparability, if CYP3A5 expression and age were adjusted for the analysis, we rated them with 1 point for each.

Statistical Analysis
Mean differences (MD) and their corresponding 95% confidence intervals (CIs) were used to identify the differences between the POR*28 genotype and PKs of TAC, and the Z-test was performed to detect the statistically significant differences between two groups. To calculate pooled estimates, we extracted the mean and standard deviation. If the studies only reported the median and interquartile range, the formulas by Wan et al. [21] were used to estimate the mean and standard deviation. Data presented as log-transformed mean and standard deviation were converted to the raw scale using the methodology of Higgins et al. [22].
We assessed the heterogeneity across studies using the chi-square test and I 2 statistics [23], and I 2 > 50% was regarded as indicating significant heterogeneity. The fixed-effect model was used if there was no significant heterogeneity; otherwise, the random-effects model was used. When we confirmed heterogeneity, a sensitivity analysis was conducted by omitting each study in turn to assess the influence of individual studies. To detect publication bias, Begg's rank correlation test and Egger's regression test were performed using R Studio software (version 3.6.0; R Foundation for Statistical Computing, Vienna, Austria) [24,25]. As the effects of POR*28 can depend on the expression status of CYP3A5, a subgroup analysis was conducted according to CYP3A5 expression status. The meta-analysis was performed using Review Manager 5.4 (The Cochrane Collaboration, Copenhagen, Denmark). Statistical significance was defined as a p-value < 0.05.

Results
Our initial search yielded 586 studies, 501 of which remained after duplicates were removed. After excluding 451 articles based on their titles and abstracts, we assessed the full text of 50 studies. Among them, 44 studies were excluded for the following reasons: evaluating other genotypes (n = 20), not having concentration data with adjustment by body weight (n = 11), not an original article (n = 6), evaluating other outcomes (n = 5), and not able to extract data (n = 2). Finally, six studies [26][27][28][29][30][31] involving 1061 patients were included in the meta-analysis ( Figure 1). The characteristics of these studies are summarized in Table 2. The studies were published between 2014 and 2018, and all were cohort studies. Four of the six studies were conducted with Asian populations, one with Caucasians, and the other with multiethnic groups. The mean age of the patients was 43.3 years (range 40.0-49.5). Quality scores evaluated using the NOS ranged from 7 to 9.

Discussion
This is the first meta-analysis investigating the association between the POR*28 polymorphism and the standardized initial trough concentration of TAC in adult renal transplant recipients. The results showed that POR*28 carriers had a lower standardized trough concentration of TAC when compared with POR*1/*1 carriers. This association was increased in CYP3A5 expressers; however, POR*28 did not affect the TAC concentration in CYP3A5 non-expressers. POR*28, a missense variant of POR, is the most common variant observed in about

Discussion
This is the first meta-analysis investigating the association between the POR*28 polymorphism and the standardized initial trough concentration of TAC in adult renal transplant recipients. The results showed that POR*28 carriers had a lower standardized trough concentration of TAC when compared with POR*1/*1 carriers. This association was increased in CYP3A5 expressers; however, POR*28 did not affect the TAC concentration in CYP3A5 non-expressers. POR*28, a missense variant of POR, is the most common variant observed in about 28% of all alleles [32]. This variant is present in the flavin adenine dinucleotide (FAD) binding site, thereby affecting POR and CYP interactions [33]. In vitro studies demonstrated that POR*28 affects CYP3A4 activity in a substrate-specific manner [34,35]. Several PK studies demonstrated that POR*28 is related to increased CYP3A activity. The study of Oneda et al. [36], which investigated CYP3A in vivo activity using midazolam, showed that POR*28/*28 was related to a 1.6-fold increase in hepatic CYP3A activity. Yang et al. [37] also showed that POR*28 was associated with increased hepatic CYP3A activity. In line with previous findings, our results regarding increased CYP3A activity might be explained by the effects of POR*28.
Several studies have reported that decreased exposure to TAC within a week after transplantation was associated with acute organ rejection. Kuypers et al. [38] reported that patients with an area under the concentration curve of 0-12 h (AUC (0-12) ) below 200 ng·h/mL had a higher risk of acute rejection when compared with those with a higher AUC (0-12) . Borobia et al. [39] also showed that patients with acute organ rejection had lower TAC trough concentrations than those without acute organ rejection. Our meta-analysis demonstrating that patients carrying the POR*28 allele had decreased TAC concentrations indicates POR*28 is an important factor in predicting acute organ rejection.
According to the subgroup analysis in this meta-analysis, POR*28 effects on the TAC concentration varied by CYP3A5 expression status, which is consistent with previous studies. For example, according to Jonge et al. [18], CYP3A5 expressers carrying the POR*28 allele required an approximately 25% higher TAC dose than CYP3A5 expressers with POR*1/*1, although the POR*28 allele did not affect TAC doses in CYP3A5 non-expressers. Gijsen et al. [40] reported that, in CYP3A5 expressers, patients with the POR*28 allele had an approximately 20% lower TAC concentration-to-dose ratio than those with POR*1/*1. However, the POR*28 polymorphism had no effect on the TAC concentration/dose ratio in CYP3A5 non-expressers. This can be explained by the role of POR, which provides electrons and enhances CYP activity.
Ethnicity may affect the expression of POR and thereby TAC metabolism. As the minor allele frequency of POR*28 was 20.0% in African Americans, 28.6% in Caucasians, and 38.9% in Asians [41], Asians are thought to be more affected by POR*28. Unfortunately, we could not compare the POR*28 effects by ethnicity, due to the small number of studies in non-Asian populations. Further studies are needed.
Our findings should be interpreted considering the following limitations. First, only six retrospective studies were included. Second, some heterogeneity existed, possibly due to the difference in the analytic methods used to determine concentrations and target concentrations. Last, although we used standardized trough concentrations after considering weight and dose, we could not adjust several factors (e.g., concurrent drugs), which can affect TAC concentrations, due to the lack of individual data.
Nevertheless, our meta-analysis demonstrated that the POR*28 polymorphism affects the TAC standardized trough concentration during the early post-transplantation period in adult renal transplant recipients, especially CYP3A5 expressers. POR and CYP3A5 genotyping might help to adjust appropriate TAC doses to reach target trough concentrations, leading to better treatment outcomes. Institutional Review Board Statement: Ethical review and approval were waived for this study due to the nature of the systematic review article.
Informed Consent Statement: Patient consent was waived due to the nature of the systematic review article.