Objectives: CYP3A5 and ABCB1 are highly implicated in the pharmacokinetics and pharmacodynamics of immunosuppressive agents, such as calcineurin inhibitors and mammalian target of rapamycin inhibitors. The polymorphisms of their coding genes play important roles in the interindividual and intraindividual differences of bioavailability of these drugs. In this study, our objective was to investigate, in a Lebanese population, the frequency of ABCB1 C3435T (rs1045642) and CYP3A5*3 (rs776746) polymorphisms and to compare the results to preexisting data from other populations. Materials and Methods: We determined the frequencies of the allelic variants of interest for 1824 Lebanese participants, and we compared these results with those from other major ethnic groups. Results: The allelic frequencies were 91.4% (C) and 8.6% (T) for CYP3A5*3 and 50.8% (T) and 49.2% (C) for ABCB1 C3435T. Our results were significantly different from most other world populations, except the European population. Conclusions: The frequencies of gene variants of interest in our Lebanese population were similar to those found in European populations. Most of our study population were CYP3A5*3 carriers, and more than half may have a lower P-glycoprotein efflux pump. These characteristics might render Lebanese transplant recipients more prone to the development of drug toxicity and in need of lower drug doses.
Key words : Calcineurin inhibitors, Drug toxicity, Mammalian target of rapamycin inhibitors, Single nucleotide polymorphisms
CYP3A5 and ABCB1 are highly implicated in the pharmacokinetics and pharmacodynamics of immunosuppressive agents, such as calcineurin inhibitors (CNIs) and mammalian target of rapamycin (mTOR) inhibitors.
CYP3A5 may be the most important genetic contributor to interindividual and interethnic differences in CYP3A-dependent drug clearance and response.1 The CYP3A5 gene is currently known to harbor at least 11 different single nucleotide polymorphisms (SNPs).1 CYP3A5*3 (rs776746) is the result of a transition (A to G), leading to a truncated nonfunctional protein2 in homozygous conditions. Accordingly, *3/*3 mutant homozygous patients are nonexpressors and thus poor metabolizers of CYP3A5 substrates, *1/*3 heterozygous patients are intermediate metabolizers, and *1/*1 wild genotype patients are normal metabolizers. The degree of expression of CYP3A5 can contribute to subtherapeutic or toxic levels of its substrates, including tacrolimus and cyclosporine. A study conducted by Coto and colleagues3 showed that the mean dose-adjusted blood tacrolimus concentration was significantly higher among CYP3A5*3 homozygous carriers than among wild-type allele CYP3A5*1 carriers.
The efflux transporter P-glycoprotein (P-gp; coded by ABCB1), the most extensively studied ATP binding cassette (ABC) transporter, also plays a major role in the pharmacokinetics of CNIs and mTOR inhibitors. It lowers intracellular concentrations of both tacrolimus and cyclosporine by pumping them out of enterocytes into the intestinal lumen.4 In addition, P-gp is involved in drug transport within hepatocytes, kidney cells, and lymphocytes.5 Interpatient variability in cyclosporine clearance could be partially explained by variations in P-gp expression.6 Over the previous decade, more than 50 SNPs have been identified in ABCB1, including the highly studied C3435T (rs1045642).7 This SNP is present at a high allelic frequency in White populations (about 50%). The C3435T TT genotype is associated with decreased expression of P-gp in renal tissue8 and thus is a risk factor for development of cyclosporine nephrotoxicity.9
The prevalence of ABCB1 and CYP3A5 polymorphisms in the Lebanese population is still unknown. Here, we investigated the frequency of CYP3A5*3 and ABCB1 C3435T in a sample of the Lebanese population. Knowledge of these polymorphisms would address the scarcity of data and assist clinicians in tailoring immunosuppressive therapy to each transplant recipient through the choice of the right drug at the appropriate dose.
Materials and Methods
This study included 1824 patients sequentially enrolled at 3 different Lebanese hospitals as part of a multicenter cross-sectional genome-wide association study (GWAS) conducted at the Lebanese American University, the Rafic Hariri University Hospital, and the “Centre Hospitalier du Nord” of Lebanon.10
This study was carried out in compliance with the Helsinki Declaration and with the approval of the Lebanese American University institutional review board. All participants signed an informed consent.
DNA was extracted from blood or buccal swabs using a standard phenol-chloroform protocol. Samples were genotyped using the Illumina Human660W-Quad BeadChip (Illumina Inc). Trained health care professionals collected further data on the sociodemographic background of all patients. The SNPs with over 98% genotyping success rate, minor allele frequency of above 1%, and in Hardy-Weinberg equilibrium (P > 1 × 10-7) were included in the analysis.
Comparison with other world populations
The genotypes of rs776746 and rs1045642 SNPs were extracted from the GWAS data using plink.11 We searched the 1000 Genome project database for frequencies of both SNPs of interest.12 We then compared the frequencies found in our sample with those of American, African, European, East Asian, and South Asian populations as listed in the database.
The statistical analysis was conducted using R software (version 3.5). The genotypes of both SNPs, as well as their respective allele frequencies, were compared using the chi-square test. P < .05 was considered to be statistically significant.
Our study included 1824 Lebanese participants: 73% were males, the mean age was 61 ± 15 years old, and the mean body mass index was 29 ± 5 kg/m2.
The allele and genotype frequencies are shown in Table 1. The genotype frequencies for CYP3A5*3 in the total study sample were 83.11% for the homozygote *3/*3 genotype, 15.96% for the heterozygous genotype *1/*3, and 0.93% for the homozygous *1/*1 genotype. In analyzed participants, the allele frequency for the polymorphic variant C was 91.4%, and the frequency of the T allele was 8.6%.
The genotype frequencies for the ABCB1 gene polymorphism C3435T in the total study sample were 26.02% for the homozygote genotype CC, 49.64% for the heterozygous genotype CT, and 24.34% for the homozygous genotype TT. In analyzed participants, the allele frequency for the polymorphic variant T was 50.8%, and the frequency of the allele C was 49.2%.
We compared the genotype/allele frequencies between our population and other populations as found in the 1000 Genome project database. We found that the frequencies in our population differed significantly from most other populations for both gene variants. The closest frequencies were found in the European population. The frequency distribution and comparison results are shown in Table 2 and Table 3 for CYP3A5*3 and ABCB1 C3435T, respectively.
We investigated the frequencies of the ABCB1 and CYP3A5 allele variants in a Lebanese population and compared these results with other ethnic groups.
For CYP3A5*3, the *1/*1 and *3/*3 genotype frequencies were considerably different from all other populations used in the comparison, especially versus the African population. Our results were closest to those found in the European population. The allele and genotype frequencies of ABCB1 C3435T in our population were similar to those for CYP3A5, that is, they were similar to the European population and significantly different from all other populations (African, American, East Asian, and South Asian).
This is not the first time that genetic profile testing in a Lebanese population showed more similarities with White populations than with populations from Middle Eastern regions. A large study conducted by Haber and associates in 2013, which investigated population relationships, showed more similarities between Lebanese and Europeans than among other world populations.13
CYP3A5 allelic frequencies are not highly studied in populations from Middle Eastern regions. In a recent study conducted on a Jordanian population, CYP3A5*3 allelic frequencies were significantly different from our results (comparison data not shown).14
For cyclosporine, published data are conflicting. Some studies have shown an association between CYP3A5*3 and the drug pharmacokinetics and dosing. For instance, Haufroid and colleagues observed a 1.6-fold higher cyclosporine dose-adjusted trough level in CYP3A5 nonexpressors among kidney transplant patients.15 Similarly, Tang and associates observed an increased dose-adjusted trough concentration for CYP3A5 nonexpressors among kidney transplant patients.16 Joy and associates conducted a study on CYP3A5 expression in the kidneys of patients with CNI nephrotoxicity. They concluded that biopsies from kidneys of patients with cyclosporine-induced nephrotoxicity have decreased expression of CYP3A5 compared with a control group.17 On another hand, some studies did not find an association between cyclosporine levels and CYP3A5*3 genotypes.14,18
Furthermore, CYP3A5 expressors (homozygous and heterozygous) were shown to have lower tacrolimus trough levels.19 According to the latest Clinical Pharmacogenetics Implementation Consortium guidelines, 3/*3 patients can be given the standard recommended dose, whereas patients carrying 1 or 2 copies of the *1 allele (ie, intermediate or normal metabolizers) should be given an increased starting dose.20
Around one-quarter of our study population were carriers of the ABCB1 3435TT genotype. Lebanese participants that belong to this group could respond differently to CNIs. The ABCB1 3435T SNP has been extensively studied to assess its effect on cyclosporine pharmacokinetics, but the results have been controversial. A meta-analysis in 2008 failed to demonstrate a definitive correlation between ABCB1 C3435T and altered pharmacokinetics of cyclosporine.21 However, new studies have shown a significant effect of the C3435T genotype with regard to cyclosporine exposure after kidney transplant.22,23
With regard to tacrolimus, some studies have suggested that individuals with ABCB1 3435 TT genotype absorb tacrolimus better and need lower daily doses.24,25 In 2019, Yildirim and associates could not find an effect of 3435T on tacrolimus dosing in their population.26 A recent meta-analysis published in 2019 could not conclude to an effect on tacrolimus pharmacokinetics.27
This was a retrospective analysis comprising a cohort of nontransplant patients. However, the large cohort size provided robust results that truly reflected the distribution of these gene polymorphisms in the general population.
This is the first study that investigated the frequency of gene variants occurring in CYP3A5 and ABCB1 in the Lebanese population. We demonstrated that the vast majority of our study population were CYP3A5*3 carriers. Furthermore, nearly one-half carried the ABCB1 3435T allele. A prospective assessment of the association between these and other genotypes and CNI exposure is of great interest to conclude on the need of genotyping in certain patients to optimize therapy and improve patient safety. If confirmed, our findings mean that Lebanese transplant recipients having high immunosuppressive drug levels or experiencing drug-induced toxicity may benefit from genotype testing if available.
Volume : 19
Issue : 5
Pages : 434 - 438
DOI : 10.6002/ect.2021.0101
From the 1School of Pharmacy and the 2School of Arts and Sciences, Lebanese American University, Byblos, Lebanon; the 3School of Medicine, Lebanese University, Beirut, Lebanon; the 4School of Medicine, Lebanese American University, Beirut, Lebanon; and the 5Rafik Hariri University Hospital, Beirut, Lebanon
Acknowledgements: The authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no declarations of potential conflicts of interest. *Pierre Zalloua and Antoine Barbari contributed equally to the study.
Corresponding author: Antoine Barbari, Rafik Hariri University Hospital, Beirut, Lebanon
Table 1. Allele and Genotype Frequencies of ABCB1 C3435T and CYP3A5*3 in a Cohort of Lebanese Population
Table 2. Comparison of CYP3A5*3 Allele (rs776746) and Genotype Frequencies Between Our Population and Other Populations as Reported by 1000 Genome Project Database