Effect of genetic polymorphisms on the pharmacokinetics of gefitinib in healthy Chinese volunteers

Abstract Gefitinib is the first-generation drug of epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) metabolised by the cytochrome P450 and transported by P-glycoprotein (ABCB1) and breast cancer resistance protein (ABCG2). In the present study, the pharmacokinetics of gefitinib in healthy Chinese volunteers was investigated and the effect of genetic polymorphisms on its variability was evaluted. Forty-five healthy volunteers were administered a single dose of gefitinib and the blood samples were used for quantifying the concentration of gefitinib and genotyping fifteen single-nucleotide polymorphisms of cytochrome P450 enzymes (CYP3A4, CYP3A5, CYP2D6, CYP2C9 and CYP2C19) and drug transporters (ABCB1 and ABCG2). CYP3A5*3 (rs776746) polymorphism showed a significant influence, with higher gefitinib AUC0-t in carrier of CC genotype than in CT/TT genotype (BH-adjusted p value <0.05). For CYP2C9*3 (rs1057910), significant differences in pharmacokinetics of gefitinib were detected between carriers of AA and AC genotypes, with higher AUC0-t, AUC0-∞ and Cmax in carrier of AC genotype than in AA gen-otype (BH-adjusted p value <0.05). No associations were found between SNPs in CYP3A4, CYP2D6, CYP2C19, ABCB1, ABCG2 and the pharmacokinetics of gefitinib. The SNPs in CYP3A5*3 (rs776746) and CYP2C9*3 (rs1057910) were found to be associated with altered gefitinib pharmacokinetics in healthy Chinese volunteers.


Introduction
Gefitinib is the first generation drug of epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) and is used for first-line treatment for the patients with advanced nonsmall cell lung cancer (NSCLC) (Hsiue et al. 2016;Rawluk and Waller 2018).Pharmacokinetic studies on healthy volunteers and patients with advanced NSCLC with EGFR mutation have showed that gefitinib has a good oral bioavailability and an extensive distribution (Mizoguchi et al. 2016;Zhang et al. 2018).The maximum plasma concentration (C max ) of the gefitinib is typically reached within 3-7 h after oral administration.The plasma elimination half-life (T 1/2 ) is 20-60 h (Zhao et al. 2017).Gefitinib exhibits a linear pharmacokinetic process within the administration range of 50-250 mg, but there are significant individual differences in pharmacokinetic parameters.The area under the concentration-time curve from 0 to infinity (AUC 0-1 ) and peak plasma concentration (C max ) by oral dosing show approximately up to 15-fold variability among interindividuals, and up to 2-fold variability among intraindividuals (Swaisland et al. 2006).
The single-nucleotide polymorphisms (SNPs) of the cytochrome P450 (CYP) enzymes and transporters are important reasons for the pharmacokinetic differences of gefitinib among NSCLC patients (Sausville et al. 2019;Wan et al. 2020).A large number of studies have reported that SNPs of CYP and the ATP-binding cassette (ABC) transporters play an important role in the safety and effectiveness of gefitinib.Gefitinib is mainly metabolised by CYP3A4, and partially metabolised by CYP3A5 and CYP2D6 (Li et al. 2007;Semba et al. 2020).Gefitinib is also the substrate of the P-glycoprotein (ABCB1) and breast cancer resistance protein (ABCG2).It has reported that ABCB1 and ABCG2 gene polymorphisms have significant effects on the therapeutic effect and adverse reactions of gefitinib (Cusatis et al. 2006;Tamura et al. 2012;Kobayashi et al. 2016).The gene polymorphisms of drug metabolism enzymes and transporters may affect the individual differences of gefitinib, but the pharmacogenetics of gefitinib remains controversial.For healthy Chinese male subjects, Ma et al. reported the impact of SNPs in the metabolising enzymes (CYP) and transporters (ABCB1 and ABCG2) on gefitinib disposition.The results suggested that a single-nucleotide polymorphism in ABCG2 (c.421C > A) significantly affected the pharmacokinetics of gefitinib, and there was no association between the pharmacokinetic parameters of gefitinib and polymorphisms in CYP3A4, CYP3A5 and CYP2D6.(Ma et al. 2019).In another research, CYP2D6 rs1058164 was associated with gefitinib exposure and may contribute to the high inter-subject variability (Zhang et al. 2018).For NSCLC patients, Ma et al. reported that the blood trough concentration of gefitinib was significantly different between CYP3A4 � 18B (rs2242480) and ABCG2 rs2231142 in Chinese NSCLC patients (Ma et al. 2019).But no significant differences were found among pharmacokinetic parameters of gefitinib and the CYP3A4 or CYP3A5 genotyps in Japanese patients with NSCLC (Kobayashi et al. 2015).
In this study, we examined the SNPs in metabolising enzymes (CYP3A4, CYP3A5, CYP2D6, CYP2C9, and CYP2C19) and transporters (ABCB1 and ABCG2) in 45 healthy Chinese volunteers who received a single oral dose of 250 mg gefitinib tablet.Additionally, we investigated the relationship between the activities of these genes and the pharmacokinetic profile of gefitinib.

Study design
We analysed the data from a bioequivalence clinical trial comparing two formulations of 250 mg gefitinib tablets.This trial involved healthy male Chinese volunteers who received a single oral dose of gefitinib tablet under fasting condition.The clinical trial was a randomised, open-label, crossover study with two periods, two sequences, and a single centre.This study was separated by a 16-day washout period, and the plasma concentrations of gefitinib were determined blindly.The enrolled 46 healthy male volunteers were all healthy Chinese males of Han ethnicity, aged 18 years or older, with a body weight not less than 50 kg and a body mass index (BMI) ranging from 19 to 26 kg/m 2 .Based on clinical history, vital sign examinations, physical examinations, clinical laboratory tests, chest X-ray and 12-lead electrocardiograms (ECGs), it was established that all subjects were healthy.
Subjects having any organic mental or psychiatric conditions, a history of drug or food allergies, or a known allergy to the ingredients of the drug were excluded.All participants were non-smokers and had refrained from consuming coffee, tea, and alcohol for a minimum of one week before the study.No additional medications were administered to the participants throughout the clinical trial, and they had the liberty to withdraw from the study whenever they wished.
One subject was excluded from the clinical trial due to diarrhoea and vomiting following gefitinib administration, and was found unfit to continue.This adverse drug reaction was classified as 'possibly related'.

Ethics statement
The protocol for this clinical study was approved by the Ethics Committee of Huzhou Central Hospital, Zhejiang, China (Approval No: 2019-015-01, 2020-019-03), and the study was conducted in accordance with the Good Clinical Practice outlined by the National Medical Products Administration, the guidelines of the Helsinki Declaration of 2013 and relevant domestic laws and regulations (ChiCTR2300074939, ChiCTR2200058996).Participants were enrolled in the study only after obtaining written informed consent.Demographic information of each subject was obtained and recorded anonymously.

Determination of plasma gefitinib concentration and its pharmacokinetic analysis
A validated high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) bioanalytical method was uesd to quantify the concentration of gefitinib in plasma (Zhang et al. 2018).Noncompartmental analysis of the pharmacokinetic parameters, including the half-life of gefitinib (T 1/2 ), T max , maximum plasma concentration (C max ), area under the concentration-time curve from 0 to 144 h (AUC (0- 144 h) ), AUC from 0 h to infinity (AUC (0-1) ), volume of distribution(Vd) and plasma clearance (CL/F) was conducted by Phoenix WinNonLin 8.0 (Certara, Princeton, New Jersey).

Statistical analysis
Statistical analysis was performed with the SPSS 22.0 software (SPSS Inc., Chicago, IL, USA).The demographic characteristics and pharmacokinetic data were expressed as the mean ± standard deviations (SD).The Hardy-Weinberg equilibrium was estimated for all analysed variants using Fisher's exact test.The parametric univariate analysis (t-test) was used to compare the pharmacokinetic parameters of gefitinib between individuals with diverse genotypes.The p values were adjusted according to the method of Benjamini/ Hochberg (B/H) (Benjamini and Hochberg 1995) to control the false discovery rate (FDR).An association was considered to be statistically significant, if its corresponding B/Hadjusted p value was below 0.05, corresponding to an FDR of 5%.Chi-square test was used to compare the incidence of ADRs between different genotypes.P values < 0.05 were considered statistically significant.

Demographic and genotypic characteristics of the subjects
From 111 volunteers screened for enrolment in the study, 45 healthy volunteers were enrolled in this study.Exclusion criteria were applied to subjects who failed to meet the inclusion criteria (n ¼ 65) or experienced an adverse drug reaction (n ¼ 1).The demographic characteristics of the included subjects are summarised in Table 1.
The genotype frequencies of enzyme and transporter genes evaluated were shown in Table 2.All genetic variants were in Hardy-Weinberg equilibrium (p � 0.05), except CYP2D6 � 10 (rs1065852) and ABCB1 rs1128503 polymorphism.The genotype of CYP2C9 � 2 (rs1799853) was CC and the genotype of ABCG2 rs72552713 was GG.So these four SNPs were excluded from the effects of polymorphism on gefitinib pharmacokinetics.
The unadjusted p values (< 0.05) of the pharmacokinetic parameters of gefitinib between individuals with diverse genotypes were adjusted according to the method of B/H (Table 5).CYP3A5 � 3 (rs776746) polymorphism showed a significant influence, with higher gefitinib AUC 0-t in carrier of CC genotype than in CT/TT genotype (BH-adjusted p value <0.05).For CYP2C9 � 3 (rs1057910), significant difference in gefitinib AUC 0-t , AUC 0-1 and C max were detected between carriers of AA and AC genotypes, with higher AUC 0-t, AUC 0-1 and C max in carrier of AC genotype than in AA genotype(BH-adjusted p value <0.05).The significant influences between gefitinib pharmacokinetic parameters and 4 SNPs were shown in Figure 1.
No significant association was identified between the studied polymorphisms and the incidence of ADRs (Table 6).However, we observed a tendency for individuals with higher AUC to experience a greater number of ADRs (CYP3A4 � 18B < rs2242480> CC and CYP2C9 � 3 <rs1057910> AC) and higher C max (CYP2C9 � 3 <rs1057910> AC).Moreover, individuals who experienced ADRs showed shorter T 1/2 (33.74 h versus 40.02 h, p ¼ 0.042) when compared to those without any ADR.However, no significant differences were observed in C max and AUC.

Discussion
Gefitinib, as an EGFR-TKI inhibitor, has significant therapeutic effect on advanced NSCLC patients with EGFR gene mutation.At standard dosage, gefitinib exhibits significant differences in pharmacokinetics among patients, which often accompanies by different therapeutic effects and adverse reactions.Several previous studies have shown that the gene polymorphism of metabolic enzymes and transporters is important for the pharmacokinetic differences among NSCLC patients.As an important metabolic enzyme in the liver, CYP3A4 plays a key role in the metabolism of gefitinib.Ma  et al. reported that CYP3A4 � 18B (rs2242480) polymorphism was associated with rash and severe abdomen in NSCLC patients with oral standard dose of gefitinib (Ma et al. 2017).
Gefitinib is also ametabolized by CYP2D6 in vivo and transforms into the active metabolite O-demethylated gefitinib.
The formation of O-demethylated gefitinib depends on the genetic polymorphisms of CYP2D6, CYP2D6 � 5 and CYP2D6 � 10, and significantly affects the clearance rate of gefitinib (Fang et al. 2017;Kwok et al. 2022).ABCB1 plays an important role in the distribution and transport of small molecule TKI drugs, and its transport ability directly affects the therapeutic efficacy and adverse reactions of TKI drugs in vivo.Ma et al. reported that ABCB1rs1128503 TT has a significant correlation with the frequency of rash and diarrhoea of gefitinib in Chinese NSCLC patients (Ma et al. 2017).ABCG2 is an important member of the ABC transporter family.ABCG2 gene polymorphism leads to different protein expression levels and changes in drug transport capacity, resulting in pharmacokinetic differences in gefitinib (Ma et al. 2019).Sakamoto et al. reported that there was a significant correlation between the ABCG2 421 C > A (rs2231142) gene mutation and gefitinib induced diarrhoea (Sakamoto et al. 2020).The inherent gene deletion led to the change of ABCG2 expression, which caused the increase of gefitinib steady-state exposure dose, and increased the risk of gefitinib induced drug-related toxicity.
In this study, we examined the association between SNPs of metabolising enzymes (CYP3A4, CYP3A5, CYP2D6, CYP2C9, and CYP2C19) and transporters (ABCB1 and ABCG2) and pharmacokinetic parameters of gefitinib in healthy Chinese volunteers who received a single oral dose of 250 mg gefitinib tablet.We found that CYP3A5 � 3 (rs776746) and CYP2C9 � 3 (rs1057910) polymorphism showed a significant influence in gefitinib pharmacokinetic parameters in healthy Chinese male subjects.Ma et al. reported that in advanced NSCLC patients, C trough was significantly lower in CYP3A4 � 18B (rs2242480) CC/CT genotype than in TT genotype (Ma et al. 2019).However, our results showed that the AUC 0-t and AUC 0-1 values of gefitinib in volunteers carrying the T allele were significantly lower than those in volunteers with the CYP3A4 � 18B (rs2242480) CC genotype when the p values were unadjusted.But no significant differences were found among pharmacokinetic parameters of gefitinib and the CYP3A4 or CYP3A5 genotyps in Japanese patients with NSCLC (Kobayashi et al. 2015).Previous studies have demonstrated that in healthy Chinese individuals, the SNP in ABCG2 rs2231142 has a significant impact on the pharmacokinetics of gefitinib (Kwok et al. 2022).However, our findings indicated no association between the pharmacokinetic parameters of gefitinib and polymorphisms in ABCG2 rs2231142, ABCG2 rs2231137, and ABCG2 rs7699188.
Even though no significant association was identified between the studied polymorphisms and the incidence of ADRs, we observed a tendency for individuals with higher AUC to experience a greater number of ADRs (CYP3A4 � 18B < rs2242480> CC, CYP3A5 � 3 <rs776746> CC and CYP2C9 � 3 <rs1057910> AC) and higher C max (CYP2C9 � 3 <rs1057910> AC) of gefitinib.Ma et al. reported that CYP3A4 � 18B (rs2242480) was marginally associated with skin rash and severe diarrhoea and CYP3A5 � 3 (rs776746) was associated with severe diarrhoea and hepatic toxicity in NSCLC patients treated with gefitinib (Ma et al. 2019).But there was no report about the association between the pharmacokinetic parameters of gefitinib and polymorphism in ABCG2 rs2622604.These inconsistent results need to be evaluated in other studies in order to elucidate whether these or other polymorphisms in ABCG2 can alter the metabolism, elimination and, as a consequence of those, the risk of suffering an ADR.We found that the degree of the effects on the pharmacokinetics and ADRs of gefitinib by the SNPs were only form 20% to 40% in healthy Chinese volunteers.Dur to the highly variable pharmacokinetic parameters of gefitinib and the small sample size in our study, further researh must be conducted to confirm the the clinical effects of the function of the metabolising enzyme and transporters gene polymorphisms.

Table 1 .
Demographic characteristics and laboratory indexes of all subjects (n ¼ 45).

Table 2 .
Genotype frequencies of enzymes and transporters genes in the individuals enrolled in the study (n ¼ 45).

Table 3 .
Pharmacokinetic variables of a single dose of 250 mg gefitinib in 45 healthy Chinese subjects with cytochrome P450 variant.

Table 4 .
Pharmacokinetic variables of a single dose of 250 mg gefitinib in 45 healthy Chinese subjects with drug transporter variants.

Table 6 .
Association between polymorphisms and adverse drug reactions(ADRs).

Table 5 .
The BH-adjusted p values of the pharmacokinetic parameters of gefitinib between individuals with diverse genotypes (unadjusted p value< 0.05).Relationship between pharmacokinetic parameters of gefitinib and CYP3A4, CYP3A5, CYP2C9 and ABCG2 genotypes.