Factors Influencing Blood Concentration of Voriconazole and Therapeutic Drug Monitoring in Patients with Child–Pugh Class C Cirrhosis

Abstract

What Is Known and Objective. CYP2C19 is an important influencing factor for voriconazole trough plasma concentration (C_min); however, it is not verified in Child–Pugh C (CP-C) cirrhosis patients, and no voriconazole dosage regimen is recommended for these patients in the package insert. This retrospective study identified CYP2C19 and other factors influencing voriconazole C_min for CP-C cirrhosis, and obtained an appropriate method of application of voriconazole for them. Methods. A total of 66 patients with CP-C cirrhosis who accepted voriconazole therapy were involved. The voriconazole C_min, clinical characteristics, CYP2C19 genotype, and adverse effects (AEs) were recorded and analyzed. Results. Unlike other research studies, voriconazole C_min was not different among normal metabolizers (NMs), intermediate metabolizers (IMs), and poor metabolizers (PMs) of the CYP2C19 enzyme in CP-C cirrhosis ( > 0.05). The maintenance dose regimen for voriconazole was the only independent influencing factor for C_min ( = 0.045; OR = 3.753; 95% CI, 1.029–13.694). At about 1/3 of the recommended maintenance dose, only 16.7% (8/48) had C_min >5.5 μg/mL, 4.5% (3/48) had C_min <1 μg/mL, and only one AE happened. There were four voriconazole-related AEs that happened in this study, and three AEs occurred (3/4, 75%) when the maintenance dose was not adjusted with therapeutic drug monitoring (TDM). What Is New and Conclusion. Voriconazole C_min did not significantly vary according to CYP2C19 enzyme metabolization status (being an NM, IM, or PM) in CP-C cirrhosis. Reducing the maintenance dose of voriconazole to approximately 1/3 the standard maintenance dose and administering in combination with TDM in patients with CP-C cirrhosis are recommended.

1. Introduction

Owing to several factors, including dysbacteriosis, the use of adrenal cortex hormone and broadspectrum antibiotics, cirrhosis-associated immune dysfunction, and invasive manipulation, invasive fungal infections (IFIs) are increasingly recognized in decompensated hepatic cirrhosis patients [1]. IFIs, in particular invasive aspergillosis (IA), constitute an important cause of morbidity and mortality [1, 2]. Voriconazole is a triazole antifungal drug that has broad antifungal effects against most yeast and mold species. Current guidelines recommend voriconazole as the first-line treatment for IA [3]. However, this drug has a narrow therapeutic range. Its established trough plasma concentration (C_min) is 1–5.5 μg/mL [4]. Some studies [5–7] have observed that under- and overdosing of voriconazole influence the efficacy and safety of therapy, respectively. Therefore, it is important to maintain the C_min in the therapeutic range.

The CYP2C19 genotype is a critical factor for determining C_min. Its variants can be classified into five subgroups: ultrarapid metabolizers (UMs) of the CYP2C19 enzyme, carriers of two CYP2C1917 (CYP2C1917/17); rapid metabolizers (RMs), carriers of one wild-type allele and one CYP2C1917 (CYP2C191/17); normal metabolizers (NMs), carriers of two wild-type alleles (CYP2C191/1); intermediate metabolizers (IMs), carriers of one null allele and one wild-type allele (CYP2C191/2 or CYP2C191/3) or carriers of one null allele and one CYP2C1917 (CYP2C192/17 or CYP2C193/17); and poor metabolizers (PMs), carriers of two null alleles (CYP2C192/3 or CYP2C192/2 or CYP2C193/3) [8, 9]. UMs and RMs carrying the CYP2C1917 allele show increased enzyme activity relative to NMs. Conversely, IMs and PMs have one or two loss-of-function CYP2C192 and 3 alleles and show significantly reduced enzyme activity, which results in higher C_min relative to NMs [10].

Nongenetic influencing factors include liver function, age, sex, route of administration (oral or intravenous), and drug-drug interactions [11]. The influencing factor for daily dose of voriconazole is disputed, and some studies [5, 11] consider that the relationship between the daily dose and C_min is weak. However, other studies [7, 12] have reported that voriconazole C_min increases with dose escalation. Therapeutic drug monitoring (TDM) is used to optimize drug therapy, which is recommended by Infectious Diseases Society of America (IDSA) for azole-based therapy of IA [3]. Due to the complex and highly variable metabolism observed with voriconazole, under- and overdosing frequently occurs, resulting in therapeutic failure or adverse events (AEs). TDM is, therefore, fundamental to maintaining voriconazole dosage in the target range [4, 13].

Liver is the main site of voriconazole metabolism [14]. Due to insufficient metabolism, voriconazole C_min is higher in patients with liver cirrhosis [15], which increases the risk for AEs, such as neurotoxicity, visual disturbance, skin reaction, and hepatic toxicity [5, 16]. Child–Pugh class C (CP-C) cirrhosis is the most serious class in the Child–Pugh classification of liver cirrhosis, causing terrible damage to liver function and critical complications such as ascites, encephalopathy, and kidney injury or even kidney failure [17]. For reasons such as multiple organ dysfunction syndrome, the use of blood purification, and drug-drug interactions, the metabolism of voriconazole and the original influencing factors of C_min may change in CP-C cirrhosis patients. Thus, we performed this study to observe voriconazole C_min, AEs, and the effect of TDM and to investigate the influencing factors, especially CYP2C19, for voriconazole C_min in such patients. Furthermore, as the dosage regimen is not mentioned in the voriconazole package insert for CP-C cirrhosis patients, this study sought clues to help inform the appropriate administration strategy for voriconazole in such patients.

2. Methods

2.1. Data Sources

We conducted a retrospective study on CP-C cirrhosis patients who diagnosed IFIs and treated with voriconazole during their hospital stay at the Fifth Medical Center of Chinese PLA General Hospital between 2014 and 2017. The diagnostic criteria from the European Organization for Research and Treatment of Cancer/Mycoses Study Group and the National Institute of Allergy and Infectious Diseases of the United States/Mycosis Study Group (2008) were used to classify IFIs as proven, probable, or possible [18]. The inclusion criteria are as follows: (1) CP-C patients diagnosed with IFIs and (2) patients treated with voriconazole. Exclusion criteria were missing important or significant clinical data and pregnancy. Patient medical records were individually reviewed using a standardized data collection template. Demographic and clinical information were collected with reference to the elements that could influence C_min. A blood sample is obtained from each patient to determine the voriconazole C_min and CYP2C19 genotype when voriconazole plasma concentration was stable.

2.2. Study Design and Voriconazole Administration

Patients were allocated to three groups according to the voriconazole C_min : C_min > 5.5 μg/mL, C_min < 1 μg/mL, and C_min between 1 and 5.5 μg/mL. The influencing factors were compared among the three groups. Voriconazole was administered by intravenous or oral (gastric canal) routes. Trough blood samples were collected at least five maintenance doses after therapy right before the next voriconazole dose. All C_min values were measured using a validated high-performance liquid chromatography (HPLC) assay described previously [19].

2.3. Ethics Approval and Consent to Participate

The study protocol and all amendments were approved by the Ethics Committee of the Fifth Medical Center of Chinese PLA General Hospital. Reviews were conducted according to the general principles outlined in the International Ethical Guidelines for Biomedical Research Involving Human Subjects, ICH Guidelines for Good Clinical Practice, and the Declaration of Helsinki. Written informed consent to participate in this study was obtained from the patients or their legal guardians/next of kin.

2.4. Data Collection

The potential influencing factors for voriconazole C_min included demographic information, such as age, sex, nationality, and body mass index (BMI); clinical characteristics, including liver and kidney function, voriconazole dosage, and delivery route; and concomitant medications that may interact with voriconazole. Liver and kidney data included total bilirubin (TBIL), direct bilirubin (DBIL), aspartate aminotransferase (AST), alanine aminotransferase (ALT), prothrombin activity (PTa), international normalized ratio (INR), albumin (ALB), model for end-stage liver disease (MELD) score, and serum creatinine (SCr). All of these data were recorded on the first day of using voriconazole. If patients had kidney or liver failure, the organ support therapies included hemodialysis (mainly continuous renal replacement therapy) and artificial extracorporeal liver support (mainly plasma exchange).

CYP2C19 genotypes were tested by BGI (Beijing, China). DNAs were extracted from 66 blood samples using the TIANamp Genomic DNA Kit (DP304, TIANGEN, Beijing, China) according to the manufacturer’s recommendations. Single-nucleotide polymorphism genotyping (including CYP2C192317) was performed by combining amplification refractory mutation system technology with the TaqMan probe.

AEs related to voriconazole were defined according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) Version 5.0 grade scores [20]. Voriconazole-related hepatotoxicity was defined by the following criteria: ALT or AST increase > upper limit of normal (ULN) when the baseline was normal or >1.5× baseline when the baseline was abnormal; alternatively, alkaline phosphatase (ALP) > ULN when the baseline was normal or >2.0× baseline when the baseline was abnormal. Other causes of hepatic deterioration were excluded, such as progression of primary liver disease, hepatic ischemic/hypoxia, and sepsis. The severity of AEs was classified according to NCI-CTCAE grades (Grade 1 = mild reaction; Grade 2 = moderate reaction; Grade 3 = severe reaction; Grade 4 = very severe reaction).

2.5. Statistical Analysis

Continuous variables are presented as either means and standard deviations (SDs) or medians with interquartile ranges (IQRs), as appropriate. An unpaired two-tailed Student's t-test or the Mann–Whitney test was used to assess the associations between continuous variables. The χ² or Fisher’s exact test when appropriate was used to assess the associations among categorical variables. Univariate analysis was performed to assess the relationship between a possible influencing factor and voriconazole C_min. Factors with  < 0.2 were entered into a logistic regression analysis.  < 0.05 was considered statistically significant. All analyses were performed with the use of SPSS software for Windows (version 23.0) and GraphPad Prism version 8.0.2.

3. Results

3.1. Patient Characteristics

In all, 66 patients were included in the study, 52 males and 14 females. All patients had CP-C cirrhosis. Voriconazole was administered to patients with proven or probable or possible IFIs. Demographic information, cause of liver injury, CYP2C19 genotype, and site of IFIs are given in Table 1. 28 patients (28/66, 42.4%) were NMs, 25 (25/66, 37.9%) were IMs, and 13 (13/66, 19.7%) were PMs. None of the patients were UMs or RMs. Each patient had at least one C_min test, and about 44% (29/66) patients had 2–5 times C_min examination. In the first test of C_min, there were 3 with C_min < 1 μg/mL (0.34 μg/mL, 0.42 μg/mL, and 0.52 μg/mL, respectively); 48 patients’ C_min were on target (mean ± SD, 3.61 ± 1.04 μg/mL); 15 patients had C_min > 5.5 μg/mL (7.52 ± 1.50 μg/mL); and there was only 1 patient with C_min > 10 μg/mL (11 µg/mL) in this group of 15. As the group of C_min < 1 μg/mL was quite small (n = 3), we combined it with the group with C_min 1–5.5 μg/mL, called C_min ≤ 5.5 μg/mL. Then, the factors were compared between groups of C_min ≤ 5.5 μg/mL and C_min > 5.5 μg/mL. The flow chart of the patients enrolled in this study is shown in Figure 1.

Table 1 

Demographic and clinical characteristics of 66 patients with Child–Pugh class C cirrhosis and indications for voriconazole therapy.

Figure 1 

Flow diagram of enrolled patients.

3.2. Factors Influencing C_min

There were two kinds of voriconazole dose regimens in this study: All 66 patients had the same loading dose (400 mg per 12 h on day 1), and then, a maintenance dose of 100 mg (1.56 ± 0.31 mg/kg) was administered twice daily for 48 patients (Regimen A) and 200 mg (3.02 ± 0.64 mg/kg) was administered twice daily for 18 patients (Regimen B). Concomitant medications mainly involved steroids and proton pump inhibitors. Univariate analysis showed that the influencing factors of voriconazole C_min were SCr level, use of hemodialysis, and voriconazole maintenance dosage schedule (Table 2). However, in multivariate analysis, the voriconazole maintenance dose schedule was identified as the only independent influencing factor for C_min (P = 0.045; OR = 3.753; 95% CI, 1.029–13.694), as shown in Table 3.

Table 2 

Factors influencing voriconazole trough plasma concentrations (C_min).

Table 3 

Results of logistic regression analysis of the influencing factors for trough plasma concentrations (C_min) of voriconazole.

3.3. Voriconazole Maintenance Dose Regimens

As the voriconazole maintenance dose schedule was the only independent influencing factor for C_min in CP-C cirrhosis patients, we further dug out the characteristics of the two kinds of dose regimens. For the 48 patients in Regimen A, the mean ± SD of C_min was 3.84 ± 1.78 μg/mL. There were 37 (37/48, 77.1%) patients whose C_min was 1–5.5 μg/mL and 8 (8/48, 16.7%) patients whose C_min was >5.5 μg/mL. 3 (3/48, 6.2%) patients had C_min < 1 μg/mL. The mean ± SD of C_min was 5.73 ± 2.53 μg/mL in Regimen B. In this group, there were 11 (11/18, 61.1%) patients whose C_min was 1–5.5 μg/mL and 7 (7/18, 38.9%) patients whose C_min was >5.5 μg/mL. There were no patients in this group with C_min < 1 μg/mL. The distribution of voriconazole C_min in Regimens A and B is given in Figure 2.

(a)

(b)

Figure 2 

Distribution of voriconazole trough plasma concentrations (C_min) in two maintenance dose regimens.

3.4. CYP2C19 Genotype

The average C_min values of NMs, IMs, and PMs were 4.34 ± 2.12 μg/mL, 4.40 ± 2.29 μg/mL, and 4.30 ± 2.14 μg/mL, respectively. Voriconazole C_min did not significantly differ among the three groups (P = 0.990). All three groups had one sample with <1 μg/mL (3.6% in NMs vs. 4% in IMs vs. 7.7% in PMs). The proportions of samples with on-target levels were 75%, 76%, and 61.5%, and those of samples with high levels were 21.4%, 20%, and 30.8%, for NMs, IMs, and PMs, respectively (Figure 3). No differences reached statistical significance among the proportion of samples for NMs, IMs, and PMs (P = 0.686).

(a)

(b)

Figure 3 

Distribution of voriconazole trough plasma concentrations (C_min) in normal metabolizers (NMs), intermediate metabolizers (IMs), and poor metabolizers (PMs) of the CYP2C19 enzyme.

In Regimen A, the proportion of NMs, IMs, and PMs was 18 : 20 : 10. The average C_min values of NMs, IMs, and PMs were 3.84 ± 1.90 μg/mL, 3.80 ± 1.57 μg/mL, and 3.90 ± 2.13 μg/mL, respectively. In Regimen B, the proportion of NMs, IMs, and PMs was 10 : 5 : 3. The average C_min values of NMs and IMs were 5.25 ± 2.31 μg/mL and 6.77 ± 3.33 μg/mL, respectively. The C_min of 3 patients of PMs was 3.39 μg/mL, 7.09 μg/mL, and 6.35 μg/mL, respectively. Voriconazole C_min was statistically insignificant among NMs, IMs, and PMs both in Regimen A and Regimen B (P = 0.990 and 0.631, respectively). The fraction of NMs, IMs, and PMs was not significantly different between Regimen A and Regimen B (P = 0.410).

3.5. AEs in Voriconazole Therapy

According to the NCI-CTCAE classification, there were four (4/66, 6.1%) voriconazole-related AEs in the study. The characteristics and the clinical course of AEs in the four patients are presented in Table 4. Patient 1 had NCI-CTCAE grade 2 hepatotoxicity (moderate), and patients 2 and 3 had grade 1 hepatotoxicity (mild). Patient 4 had grade 2 consciousness disturbance (moderate).

Table 4 

Characteristics of and clinical course in patients with voriconazole-related adverse events (AEs).

All four AEs had higher C_min, and 3 AEs had voriconazole C_min > 5.5 µg/mL. One AE occurred in Regimen A (1/48, 2.1%), and three occurred in Regimen B (3/18, 16.7%); however, there were no significant differences between Regimen A and Regimen B (χ² = 2.664, P = 0.103). Of the 15 patients with C_min > 5.5 μg/mL, 11 received adjusted voriconazole regimens after the C_min results were identified, and none experienced AEs. Four patients did not receive timely adjustments, and three of them experienced voriconazole-related AEs (P = 0.009).

4. Discussion

This retrospective study analyzed the characteristics of voriconazole C_min; in particular, the factors influencing C_min in patients with CP-C cirrhosis. To the best of our knowledge, few studies have focused on the application of voriconazole to patients with serious liver damage.

CYP2C19 genotype is a crucial determinant for voriconazole metabolism in the liver, and the distribution of its subtypes varies across ethnic populations. The 17 allele has been reported to exist more often in Caucasians (16–21%) and Africans (16%) than in Asians (3–6%), and the 2 and 3 alleles occur at higher frequencies in Asians than in Caucasians [21]. Almost all patients in this study were of the Han ethnicity (65/66) and had frequencies of CYP2C191, 2, 3, and 17 at 60.6% (80/132), 31.8% (42/132), 6.8% (9/132), and 0.8% (1/132), respectively, consistent with the literature [10, 22]. Some studies [23, 24] have demonstrated that voriconazole C_min is approximately two to four times higher in PMs and IMs than in NMs. However, our data showed that C_min did not significantly vary according to metabolic phenotype (NMs, IMs, and PMs). Lin et al. [10] also did not find significant differences of voriconazole C_min among NMs, IMs, and PMs in patients with hepatic dysfunction. The research of Tang et al. [25] had the same conclusion. A possible reason for this was that the effect of the decreased activity of CYP2C19 enzyme may be more significant than the genetic polymorphism of CYP2C19 itself for patients with liver dysfunction.

CYP2C19 enzyme is part of the CYP450 system. This system exists mainly in the liver; it requires nicotinamide adenine dinucleotide phosphate (NADPH) and molecular oxygen to complete its action [26]. Due to severe portosystemic shunting, decreased liver perfusion, and sinusoidal capillarization in CP-C cirrhosis patients, CYP450 enzymes cannot obtain enough oxygen to carry out their function in such patients. Studies have confirmed that the activities of CYP450 enzymes decrease with the aggravation of liver disease, and CYP2C19 is particularly sensitive to liver damage [27, 28]. The reduced liver cell mass and decreased function of CYP2C19 could result in insufficient metabolism of voriconazole and relatively high voriconazole plasma concentrations; thus, it is necessary to reduce the maintenance dose of voriconazole in patients with severe liver cirrhosis.

A retrospective study that included six cirrhosis patients at the CP-C cirrhosis stage reported that maintenance doses of voriconazole should be reduced to approximately 1/3 of the recommended dosage regimen [29]. Ren et al. suggested that the maintenance dose of voriconazole should be approximately 1/4 of the standard dose in patients with CP-B and C cirrhosis to attain the optimal C_min [19]. In this study, when voriconazole was dosed per kg body weight, the mean maintenance dose was approximately 1/3 of the standard dose in Regimen A (1.56 mg/kg vs. 4 mg/kg) and 3/4 of the standard dose in Regimen B (3.02 mg/kg vs. 4 mg/kg), respectively. Clearly, Regimen B was inappropriate for patients with CP-C cirrhosis because the mean ± SD of C_min was 5.73 ± 2.53 μg/mL, and 38.9% (7/18) patients had C_min > 5.5 μg/mL in this group. But in Regimen A, the mean ± SD of C_min was 3.84 ± 1.78 μg/mL, and only 16.7% (8/48) of patients had C_min > 5.5 μg/mL. This was still a little higher than voriconazole used as the standard maintenance dosage regimen in two previous studies on patients with normal liver function (10–12%) [30, 31]. However, in those studies, there were 18–20% of patients with voriconazole C_min < 1 μg/mL, which would be associated with therapeutic failure, while there were only three patients with C_min < 1 μg/mL (4.5%, 3/66) in Regimen A. Efficacy should be the main aim in therapy to ensure success. A regimen of less than 1/3 of the maintenance dose could be safer, but it would entail the greater likelihood of C_min < 1 μg/mL.

In addition, the mean voriconazole C_min of Regimen A was 3.84 μg/mL, higher than the C_min in normal liver function [31]. We consider that this could lead to a good outcome for the patients because higher C_min values signify better therapeutic effects [5, 11, 32], which is more important for treating lethal fungal infection. A previous study also found that a mean voriconazole C_min of 3.0–4.0 μg/mL had a maximum clinical response for yeasts and molds (81% and 74%, respectively) [33].

However, taking into account that higher voriconazole C_min could induce more toxic reactions [5, 11], AEs were carefully recorded according to the NCI-CTCAE grades; surprisingly, only four AEs (4/66, 6.1%) were observed in the study, far fewer than that observed in normal liver function populations treated with voriconazole. Dose adjustment according to TDM could be an important explanation for this. In the 15 patients with C_min > 5.5 μg/mL, 11 had adjusted voriconazole regimens or stopped voriconazole therapy according to the pharmacist’s advice, and no patient developed AEs. Four patients did not have adjusted voriconazole dosage for unknown reasons, and three of the four developed voriconazole-related AEs (hepatotoxicity). TDM is therefore recommended to identify inappropriate exposure and help clinicians adjust the voriconazole dose regimen to avoid AEs [5, 6, 11, 13, 30]. The importance of TDM in improving therapeutic safety in liver dysfunction has also been confirmed [10, 15, 29].

Although Regimen B was not appropriate, there were only three AEs that happened in this group. All three AEs were mild or moderate. The two mild hepatotoxicities could not even be discovered during voriconazole therapy. This might be one reason why the clinicians chose this kind of maintenance dose regimen. Another possible reason is that this dose regimen could quickly achieve the intended level of C_min, which is good for treating IFIs; then TDM could help adjust the voriconazole dose. It is noteworthy that only one AE happened in Regimen A. In this case, the voriconazole dose was not adjusted with TDM, which indicates that this maintenance dosage regimen coupled with TDM could be adopted for CP-C cirrhosis patients.

In our cohort, the factors indicating liver function, such as AST, TBIL, DBIL, ALB, PTa, and INR, had no positive correlation with voriconazole C_min, whereas previous studies have identified those factors could influence voriconazole C_min [31, 34, 35]. We speculated that the negative results could be due to the heterogeneity of the investigated populations or each indicator single alone could not represent the liver metabolic function of voriconazole in CP-C cirrhosis. Moreover, our data did not support that PPIs and glucocorticoids could induce significant changes in voriconazole C_min. Considering that both PPIs and glucocorticoids were able to influence voriconazole C_min by acting on CYP enzymes (PPIs as inhibitors and glucocorticoids as inducers) [11, 34, 36], the conflicting results could be explained by the complicated interaction among those comedications, voriconazole, and severe damage of CYP system in CP-C cirrhosis patients.

The present study had some limitations. First, it was a single-center, relatively small study; the results should be verified in a larger population. In addition, voriconazole was not dosed by body weight when used on real patients; a prospective study should be designed to confirm our conclusion. Finally, this study lacked UMs and RMs of the CYP2C19 enzyme, further research should address these groups.

5. What Is New and Conclusion

CYP2C19 polymorphisms were not significant predictors of voriconazole C_min in patients with CP-C cirrhosis. Due to the severe damage to the mass of hepatic cells and the decreased activity of CYP enzymes in CP-C cirrhosis patients, the metabolic clearance of the liver is greatly decreased, which can induce excessive accumulation of voriconazole and increase voriconazole plasma concentration. In this study, when the maintenance dose of voriconazole was reduced to about 1/3 of the recommended maintenance dose for patients with CP-C cirrhosis, only 16.7% of patients had C_min > 5.5 μg/mL and 4.5% of patients had C_min < 1 μg/mL. 3 AEs (3/4, 75%) happened when voriconazole C_min > 5.5 μg/mL, and TDM could be used as a routine tool to optimize voriconazole administration in such patients.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest to disclose.

Authors’ Contributions

Ying Zhang contributed to methodology, investigation, data curation, formal analysis, writing the original draft, and conceptualization of the study. Rongrong Wu was involved in conceptualization and methodology. Fangfang Liu was responsible for methodology, formal analysis, and reviewing and editing the draft. Yonggang Wang contributed to data curation and formal analysis. Chengcheng Ji, Junchang Zhang, and Xianghong Lu were involved in provision of resource and data curation. Dan Chang contributed to provision of resources and investigation. Jinsong Mu contributed to conceptualization, methodology, resources, reviewing and editing the draft.

Acknowledgments

The authors thank the participants for their involvement in this project. Special thanks go to Doctor Yuan Gao (Fourth Department of Hepatology, Beijing Youan Hospital, Capital Medical University, Beijing, China) for the important suggestions given for the revision of this paper. Thanks as well are given to Lu Chen (Beijing Macro & Micro-Test Bio-Tech Co., Ltd. Beijing, China) for consultation about the CYP2C19 genotype test.