Improvements in CYP2C9 Genotyping Accuracy Are Needed: A Report of the First Proficiency Testing for Warfarin-related CYP2C9 and VKORC1 Genotyping in China.

ORIGINAL ARTICLE

Improvements in CYP2C9 Genotyping Accuracy Are Needed: A Report of the First Proficiency Testing for Warfarin-related CYP2C9 and VKORC1 Genotyping in China Guigao Lin, MD,* Lang Yi, BS,*† Kuo Zhang, MS,* Yu Sun, MD,* Lunan Wang, PhD,* Rui Zhang, MD,* Jiehong Xie, BS,* and Jinming Li, MD*

Abstract: Warfarin is the most commonly used oral anticoagulant in clinical practice. The cytochrome P450 2C9 (CYP2C9) and vitamin K epoxide reductase complex 1 (VKORC1) genotypes have been confirmed to be associated with warfarin dose requirements. Accurate genotyping results are of particular importance for obtaining reliable genotype-guided warfarin dosing information. This study aims to determine analytic performance of laboratories offering CYP2C9 and VKORC1 testing in China. A proficiency panel of 15 validated cell samples covering common CYP2C9 and VKORC1 genetic polymorphisms was provided to 31 participating laboratories, and their genotyping results were evaluated. Fourteen data sets (45.2%) performed well with the entire panel of samples, and 17 data sets (54.8%) reported at least one genotyping error. For VKORC1 (21639G.A), participating laboratories were 100% successful in detecting genotypes of GG, GA, and AA. For CYP2C9, participants were greater than 90% successful in detecting genotypes of *1/*1, *1/*2, *1/*3, *2/*3, and *3/*3. However, 15 laboratories failed to detect rarely encountered variant genotype *2/*2. The poor performance of CYP2C9 genotyping may be because of the limitation of methodologies used for detecting CYP2C9*2 allele. The proficiency testing survey highlighted the need for improving genotyping accuracy for some laboratories in this field. Key Words: CYP2C9, genotyping, proficiency testing, VKORC1, warfarin (J Cardiovasc Pharmacol 2015;66:129–134)

Received for publication January 5, 2015; accepted March 9, 2015. From the *National Center for Clinical Laboratories, Beijing Hospital, Beijing, China; and †Graduate School, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, China. Supported by the capital development fund from the Beijing Municipal Public Health Bureau (No. 2011-4011-02) and the Special Fund for Healthscientific Research in the Public Interest from National Population and Family Planning Commission of the People’s Republic of China (No. 201402018). The authors report no conflicts of interest. G. Lin and L. Yi contributed equally to the work. Reprints: Jinming Li, MD, National Center for Clinical Laboratories, Beijing Hospital, 1 Dahua Rd, Dongdan, Beijing 100730, China (e-mail: jmli@ nccl.org.cn). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

J Cardiovasc Pharmacol Volume 66, Number 2, August 2015

INTRODUCTION Warfarin is the most widely used anticoagulant drug for prevention and treatment of thromboembolic disorders.1 However, warfarin has a narrow margin between therapeutic inhibition of clot formation and bleeding complications.2 To achieve a balance between the desired therapeutic effect and the risk of bleeding, it is necessary to identify the right dose for each patient. Dosing of warfarin is challenging because of a wide interindividual variability in patient response.3 Currently, dose adjustments are based on maintaining the international normalized ratio within a narrow therapeutic range. An increase in the international normalized ratio above the therapeutic range increases the risk of bleeding, which often results in drug-related hospitalization.4 Numerous observational studies have shown that genetic polymorphisms of cytochrome P450 2C9 (CYP2C9) and vitamin K epoxide reductase oxidase complex 1 (VKORC1) explain about one third of the variability in warfarin daily dose.5,6 S-warfarin is mainly metabolized by CYP2C9 isoenzyme, and CYP2C9*2 (rs1799853) and *3 (rs1057910) variants decrease the degradation and clearance of warfarin.7 The VKORC1 21639G.A polymorphism (rs9923231) is located in the promoter of the VKORC1 gene and could alter warfarin sensitivity and dose requirements.8 These findings have resulted in 2 updates of the US Food and Drug Administration–approved warfarin label,9 which emphasized the importance of genotype information for initial dosing estimates of warfarin. However, an ongoing controversy exists regarding the clinical utility of warfarin genotyping. The conflicting results of the 2 recent high-profile, randomized controlled trials indicate that more information is still necessary to determine the utility of genetic testing in the management of thrombosis.10–12 Proficiency testing (PT) is a system of evaluating laboratory performance objectively by an external agency. The PT process can identify systematic errors in methodology, which may not be revealed by internal quality assessment processes. To our knowledge, warfarin sensitivity genotyping has been carried out in many clinical laboratories in China. However, some of them have just launched clinical pharmacogenomics testing recently and have not taken strict clinical standards. However, a variety of methods, including polymerase chain reaction (PCR)/sequencing,13 www.jcvp.org |

129

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.


Lin et al

pyrosequencing,14 PCR–restriction fragment length polymorphism (RFLP),15 real-time PCR,16 and microarray,17 were applied to warfarin pharmacogenetic testing by the clinical laboratory community. Each method has its own benefits and limitations. Harmonization of testing methods is critical for reliable CYP2C9 and VKORC1 genotyping in clinical practice. Because the correctness of the warfarin genotyping result is very important for good patient care, the National Center for Clinical Laboratories conducted a PT survey aimed at assessing the performance of CYP2C9 and VKORC1 testing in China. Different from the College of American Pathologists pharmacogenetics survey,18 we used cell samples instead of DNA to evaluate the entire genotyping process, including DNA extraction. The proficiency panel was distributed to 31 laboratories nationwide and analyzed using different test methods in a blinded manner. Here, we provided a baseline picture of the accuracy of CYP2C9 and VKORC1 genotyping in China.

METHODS Scheme Organization Chinese laboratories that performed CYP2C9 and VKORC1 polymorphism analyses were invited to participate in the PT survey. There were no criteria set for participating. In total, 31 laboratories participated in the regional CYP2C9 and VKORC1 PT scheme in 2014. To assess the whole genotyping process, cell samples were prepared and distributed to each participant laboratory. The CYP2C9 PT panel (n = 10) consisted of 7 mutant samples and 3 wild-type samples (Table 1). The VKORC1 PT panel (n = 5) consisted of 3 mutant samples and 2 wild-type samples. The participants were assigned the same coded samples and were requested to perform DNA extraction and analysis using their routine procedures. Cell samples were shipped at ambient temperature and advised to be evaluated and processed as soon as the laboratory receives them. The laboratories were given 14 days to complete the analyses and submit the results of genotyping. All results were submitted electronically. The participants were also requested to provide information regarding the procedures used, including the methods and reagents used for DNA extraction and mutation analysis, and the concentration of extracted DNA.

Preparation of Cell Samples Cell lines (Table 1) harboring common CYP2C9 and VKORC1 polymorphisms were purchased from Coriell Cell Repositories (Coriell, NJ). The genetic polymorphisms of these cell lines have been well characterized by the Genetic Testing Reference Materials Coordination program.19 Briefly, cells were grown in Roswell Park Memorial Institute 1640 supplemented with fetal bovine serum, 2 mM L-glutamine, 10 U/mL penicillin, and 10 mg/mL streptomycin. On the day of sample distribution, cells were collected and diluted in fresh medium to a concentration of 2 · 106 cells per milliliter. Five hundred microliters of each preparation were dispensed in 1.5-mL vials and labeled.

Validation of Proficiency Panel Before distribution of the CYP2C9 and VKORC1 proficiency panels, the samples were tested using 2 methods by the National Center for Clinical Laboratories reference laboratory. PCR sequencing and PCR–RFLP assay to verify the genotypes were performed on all the cell lines. For direct sequencing, DNA was first isolated from cell samples using the QIAamp DNA Mini kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions and then amplified with specific primers for genes CYP2C920 and VKORC1.7 The PCR products were sequenced in both directions by ABI 3500DX Genetic Analyzer (Applied Biosystems, CA) and BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems) according to the manufacturer’s instructions. Genotyping by PCR–RFLP technique was conducted as previously described.15 Briefly, the extracted DNA was subjected to PCR amplification with specific primers containing restricted enzyme site and then digested using relevant restriction endonuclease (Thermo Fisher, Waltham). The digested products were analyzed by agarose gel and visualized with ethidium bromide under ultraviolet illumination. For stability study, the dispensed cell samples were incubated at room temperature for 1 week and then analyzed.

Data Analysis Genotyping results were compared with the genotype determined by the reference laboratory. At least 80% genotype accuracy is the criteria for considering a data set proficient. A PT report summarizing the data gathered and detailing issues was returned to participants, and the results of

TABLE 1. Sample Validation Coriell Cell Line Number GM17114 GM17209 GM17215 GM17222 GM17285 GM17247 GM17114 GM16654 GM17216

130

| www.jcvp.org

Coriell Characterization (Gene/Genotype)

PCR/Sequencing (Gene/Genotype)

PCR–RFLP (Gene/Genotype)

Sample Code

CYP2C9*1/*1 CYP2C9*1/*2 CYP2C9*1/*3 CYP2C9*2/*2 CYP2C9*2/*3 CYP2C9*3/*3 VKORC121639GG VKORC121639GA VKORC121639AA



C1401, C1403, C1407 C1409 C1402, C1405 C1408 C1404, C1410 C1406 V1402, V1403 V1405 V1401, V1404

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.



Quality of Warfarin-related Genotyping in China

Performance of CYP2C9 and VKORC1 Genotyping and Participating Laboratories

PT scheme were discussed with the participants on a special meeting. The results produced by testing methods were compared with the reference Sanger sequencing results to determine sensitivity, specificity, and accuracy. The agreement between genotyping methods and reference Sanger sequencing results was analyzed by the kappa index. Confidence intervals of 95% were determined for all parameters analyzed. Analyses were performed using the MedCalc software program (MedCalc Software, Mariakerke, Belgium).

Table 2 shows the results of genotype accuracy for participants in the 2014 survey. A total of 29 false-negative results21 were observed, among which 6 were that a mutation was present, but it was not reported by the participants (wild type instead of mutation). The other 23 were wrong mutations (mutation reported, but wrong). For CYP2C9 genotyping, a systematic error was found. Poor accuracy (51.6%) was reported for CYP2C9*2/*2 genotype, whereas the accuracy of other CYP2C9 genotype was greater than 90%. It is worth noting that 15 of 31 laboratories reported CYP2C9*2/*2 genotype as *1/*2. Among the 15 error responses, 12 were generated by the 12 laboratories using BaiO CYP2C9&VKORC1 genotyping kit (BaiO, Shanghai, China), which is based on the PCR–microarray method, 2 errors were produced from 2 laboratories using commercial kits based on either pyrosequencing or real-time PCR method, and 1 error was reported by a laboratory with in-house Sanger sequencing method. Table 3 presents error distribution in terms of type of laboratory. The proficiency of CYP2C9 and VKORC1 genotyping by assay is shown in Table 4. Only 45.2% of the participants correctly identified all the 15 PT challenges. The best results (correct identification of all challenges) were provided by laboratories using pyrosequencing (87.5%), real-time PCR (80.0%), and Sanger sequencing (66.7%). Fifteen data sets were 80%–99% proficient. However, 2 data sets, produced with the method of single-base extension based on ABI SNaPshot Multiplex system (Applied Biosystems), were not proficient. We next assessed the performance of different testing methodologies (Table 5). The overall performance of the assays for the CYP2C9 gene was poorer than VKORC1 gene. For CYP2C9 genotyping, most methods show good

RESULTS Sample Validation Table 1 summarizes the results of the verification. Two verifying methods gave concordant results. Stability study showed that the amount of DNA (.15 mg) extracted from each cell sample, which was incubated for 7 days, is adequate for downstream analysis.

Panel Distribution and Response All the 31 participants, including 24 hospital laboratories, 3 commercial laboratories, and 4 reagent manufacturers, submitted the results within the established deadline. Twenty-seven laboratories offer tests for clinical use. The PT samples were delivered to laboratories across the country within 7 days. None of the participants reported DNA extraction problems with the cell samples. The main methodology used by the participants was PCR–microarray (13 laboratories, 41.9%). Twenty-four laboratories used commercial kits developed by manufacturers for research purposes or for use in clinical diagnostics, and 7 participants used laboratory-developed methods.

TABLE 2. Summary of Results for Genotype Concordance for CYP2C9 and VKORC1 Sample

Gene/Genotype

No. Correct/Total

Concordance, %

No. Error

Error Response

Error Type

C1401 C1402 C1403 C1404

CYP2C9*1/*1 CYP2C9*1/*3 CYP2C9*1/*1 CYP2C9*2/*3

31/31 29/31 31/31 28/31

100 93.5 100 90.3

0 2 0 3

NA False negative NA False negative

C1405

CYP2C9*1/*3

28/31

90.3

3

C1406 C1407 C1408 C1409 C1410 V1401 V1402 V1403 V1404 V1405

CYP2C9*3/*3 CYP2C9*1/*1 CYP2C9*2/*2 CYP2C9*1/*2 CYP2C9*2/*3 VKORC121639AA VKORC121639GG VKORC121639GG VKORC121639AA VKORC121639GA

29/31 31/31 16/31 29/31 29/31 31/31 31/31 31/31 31/31 31/31

93.5 100 51.6 93.5 93.5 100 100 100 100 100

2 0 15 2 2 0 0 0 0 0

NA CYP2C9*1/*1 (n = 2) NA CYP2C9*1/*3 (n = 1) CYP2C9*1/*2 (n = 2) CYP2C9*2/*3 (n = 1) CYP2C9*1/*1 (n = 2) CYP2C9*1/*1 (n = 2) NA CYP2C9*1/*2 (n = 15) CYP2C9*1/*3 (n = 2) CYP2C9*1/*2 (n = 2) NA NA NA NA NA

False negative False negative NA False negative False negative False negative NA NA NA NA NA

NA, not applicable.


www.jcvp.org |

131



Lin et al

45.2% of the participants correctly identified all CYP2C9 and VKORC1 genetic polymorphisms. The overall genotype concordance of CYP2C9 in this PT [281 responses/310 challenges (90.6%)] is poorer than the College of American Pathologists survey [916 responses/930 challenges (98.5%)].18 However, perfect score (100%) was achieved for VKORC1 genotyping. VKORC1 genotype was reported to be the main genetic factor associated with interindividual variability in warfarin dose in Chinese population.23–25 The poor results regarding CYP2C9 genotyping may be explained by lack of a good validation of the methods used. Residual patient specimens are often used for genetic test validation; however, the allele frequency for CYP2C9*2 among Chinese population is almost zero26; thus, the mutant samples are difficult to obtain. Most of the commercial detecting kits use artificial mutant plasmid as positive controls in method validation, but the synthetic control materials do not resemble real patient specimens and may have problems in clinical tests. In this study, 12 systematic errors of reporting CYP2C9*2/*2 as *1/*2 (sample C1408) were produced by the BaiO CYP2C9&VKORC1 genotyping kit, which used constructed mutant plasmid as positive controls. The kit was able to detect single CYP2C9*2 allele but failed to identify homozygous *2 alleles. Among the 29 genotype errors identified, 18 were caused by failure detection of CYP2C9*2 allele. Generally, if the wrong mutation is reported, the resulting risk assessments will be inappropriate. Six false wild types were found. Two laboratories incorrectly reported CYP2C9*3/*3 as CYP2C9*1/*1, with the method of singlebase extension. These false-negative genotyping results will mistakenly predict warfarin sensitivity, which may incorrectly reduce the predicted over-anticoagulation and bleeding risk for patients in a warfarin anticoagulation clinic setting. Whether this failure is because of the methodology or the laboratory is not known yet. Some of the errors may be related to laboratory performance. For example, a laboratory using laboratorydeveloped method reported 6 wrong results in 15 samples. Thus, it is important for laboratories to validate laboratory-developed mutation tests before using them. Remarkably, we found 2 data sets with 6 identical genotype errors. The 2 laboratories used the same methodology, and their results might have been discussed.

TABLE 3. Error Distribution According to Type of Laboratory Laboratory

No. Error

Error Type

13 7

False negative False negative

0

False negative

1

False negative

1

False negative

6 1

False negative False negative

Medical laboratory (n = 9) Laboratory from Department of Pharmacy (n = 12) Laboratory from Department of Pathology (n = 1) Laboratory from Department of Hematology (n = 1) Laboratory from Department of Gastroenterology (n = 1) Commercial laboratory (n = 3) Reagent manufacturer (n = 4)

sensitivity, specificity, and accuracy. However, the analytical sensitivity of the in-house single-base extension assay was really low (14.2%; 95% confidence interval, 1.70%–42.8%) in the detection of CYP2C9 polymorphism. Eight CYP2C9 and VKORC1 genotyping methods, except for the in-house single-base extension assay (kappa index = 0.090), were reliable (kappa index . 0.750) in comparison with the reference Sanger sequencing (considered the gold standard) results.

DISCUSSION VKORC1 and CYP2C9 genetic polymorphisms are associated with wide interindividual variability in the metabolism of warfarin. Patients who carry VKORC1 (21639AA), CYP2C9*2, and CYP2C9*3 polymorphisms require a lower dose and have an increased risk for warfarin over-anticoagulation.22 Genotyping for VKORC1 and CYP2C9 variants can allow individualization of the dose for warfarin. However, the implementation of genotype-guided warfarin dosing will be dependent on the accuracy of genotyping. Like for any other clinical test, quality assurance is also important for VKORC1 and CYP2C9 testing, but to date, there is limited experience. To fill this gap, a PT survey was set up to assess the genotype accuracy of the 31 laboratories. Our PT results show that only

TABLE 4. Proficiency Results of Each Assay No. Data Sets Proficient at Warfarin Testing Assay All assay PCR–microarray, BaiO PCR–microarray, Daan Pyrosequencing, QIAGEN Real-time PCR, Sinomdgene Real-time PCR, Skybiotech Real-time PCR, Yzybio In-house real-time PCR In-house Sanger sequencing In-house single-base extension

No. Data Sets

100%

99%–90%

89%–80%

,80%

31 12 1 8 1 1 1 2 3 2

14 0 1 7 0 1 1 2 2 0

14 11 0 1 1 0 0 0 1 0

1 1 0 0 0 0 0 0 0 0

2 0 0 0 0 0 0 0 0 2

100% proficient, all genotype detected correctly; 80%–99% proficient, 80%–99% of genotype detected correctly; not proficient, ,80% of genotype detected correctly.

132

| www.jcvp.org




Quality of Warfarin-related Genotyping in China

TABLE 5. Performance Characteristics of Different Assays Compared With Reference Sanger Sequencing Method CYP2C9

Assay PCR–microarray, BaiO PCR–microarray, Daan Pyrosequencing, QIAGEN Real-time PCR, Sinomdgene Real-time PCR, Skybiotech Real-time PCR, Yzybio In-house real-time PCR In-house Sanger sequencing In-house single-base extension Total

No. Data Sets

Sensitivity (%; 95% CI) Correct Positive Responses/Positive Challenges

Specificity (%; 95% CI) Correct Negative Responses/Negative Challenges

Accuracy (%; 95% CI) Correct Responses/Total Challenges

Kappa Index (%; 95% CI)

12 1 8 1

83.3; 73.6–90.5 (70/84) 100; 59.0–100 (7/7) 98.2; 90.4–99.9 (55/56) 85.7; 42.3–99.6 (6/7)

100; 90.2–100 (36/36) 100; 29.2–100 (3/3) 100; 85.7–100 (24/24) 100; 29.2–100 (3/3)

88.3; 82.5–94.0 (106/120) 100; 100–100 (10/10) 98.7; 96.3–100 (79/80) 90.0; 71.4–100 (9/10)

0.750; 0.631–0.869 1; 1–1 0.971; 0.913–1 0.783; 0.388–1

1 1 2 3 2

100; 59.0–100 (7/7) 100; 59.0–100 (7/7) 100; 76.8–100 (14/14) 95.2; 76.1–99.8 (20/21) 14.2; 1.70–42.8 (2/14)

31

86.6; 81.3–90.8 (188/217)

100; 100; 100; 100; 100;

29.2–100 29.2–100 54.7–100 66.3–100 54.7–100

(3/3) (3/3) (6/6) (9/9) (6/6)

100; 96.1–100 (93/93)

100; 100–100 (10/10) 100; 100–100 (10/10) 100; 100–100 (20/20) 96.6; 90.2–100 (29/30) 40.0; 18.5–61.4 (8/20)

1; 1; 1; 0.833; 0.090;

90.6; 87.4–93.8 (281/310)

1–1 1–1 1–1 0.518–1 0–0.226

0.795; 0.726–0.865

VKORC1

Assay PCR–microarray, BaiO PCR–microarray, Daan Pyrosequencing, QIAGEN Real-time PCR, Sinomdgene Real-time PCR, Skybiotech Real-time PCR, Yzybio In-house real-time PCR In-house Sanger sequencing In-house single-base extension Total

No. Data Sets

Sensitivity (%; 95% CI) Correct Positive Responses/Positive Challenges

Specificity (%; 95% CI) Correct Negative Responses/Negative Challenges

Accuracy (%; 95% CI) Correct Responses/Total Challenges

12 1 8

100; 90.2–100 (36/36) 100; 29.2–100 (3/3) 100; 85.7–100 (24/24)

100; 85.7–100 (24/24) 100; 15.8–100 (2/2) 100; 79.4–100 (16/16)

100; 100–100 (60/60) 100; 100–100 (5/5) 100; 100–100 (40/40)

1; 1–1 1; 1–1 1; 1–1

1

100; 29.2–100 (3/3)

100; 15.8–100 (2/2)

100; 100–100 (5/5)

1; 1–1

1

100; 29.2–100 (3/3)

100; 15.8–100 (2/2)

100; 100–100 (5/5)

1; 1–1

1 2 3

100; 29.2–100 (3/3) 100; 54.7–100 (6/6) 100; 66.3–100 (9/9)

100; 15.8–100 (2/2) 100; 39.7–100 (4/4) 100; 54.7–100 (6/6)

100; 100–100 (5/5) 100; 100–100 (10/10) 100; 100–100 (15/15)

1; 1–1 1; 1–1 1; 1–1

2

100; 54.7–100 (6/6)

100; 39.7–100 (4/4)

100; 100–100 (10/10)

1; 1–1

31

100; 96.1–100 (93/93)

100; 94.2–100 (62/62)

100; 100–100 (155/155)

1; 1–1

Kappa Index (%; 95% CI)

CI, confidence interval.

Even if it occurs, interlaboratory discussion is prohibited in PT. Over time, as more laboratories in China enroll in the CYP2C9 and VKORC1 genotyping PT survey, we may provide information on possible association between genotyping methods and the number and kind of errors. An important issue in warfarin-related genotyping is the technique used for testing. Sanger sequencing is considered to be the gold standard because this methodology can identify all possible mutations in the analyzed gene segment. However, our study mainly focused on the genotyping quality of laboratories rather than choosing the best methodology. In this study, testing kits developed by in vitro diagnostic manufacturers are predominately used for CYP2C9 and VKORC1 genotyping. These kits have the advantages of being validated, ready for use, and quality controlled. Laboratory-developed tests may be less expensive but need to be validated27 and may have limited quality control. Therefore, the use of commercial kits might be Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

advisable in laboratories, which have low experience in molecular techniques. A number of new technologies have also been developed, such as high-resolution melting analysis28 and surface-enhanced laser desorption and ionization time-of-flight (SELDI-TOF) mass spectrometry.29 Currently, most genotyping technologies involve PCR amplification of target sequence. PCR is an extremely sensitive technique, which means even very low levels of contamination with the target DNA will result in a positive signal. In a diagnostic laboratory, there can be more opportunities for PCR contamination because of the repeated analyses of selected templates. In China, medical laboratories providing molecular genetic testing services for clinical purposes must be certified according to the “Regulation of Clinical Genetic Testing Laboratory.” This regulation, which focuses on contamination prevention and standard operation procedure, is issued by the government in 2002. Among the www.jcvp.org |

133



Lin et al

24 hospital laboratories participated in this PT scheme, 9 are medical laboratories and 15 laboratories are from other disciplines (Table 3). The latter ones in China have little or no previous experience in molecular genetic testing and have not given sufficient attention to issues of quality. Although falsepositive results were not found in our PT survey, the problem of PCR contamination should always be kept in mind. A PT scheme provides objective evidence of laboratory testing quality; moreover, it would be of some significance to clinicians. They may learn that quality control will have a great effect on producing reliable results in genetic testing laboratory. Under these circumstances, physicians should care not only about the testing results but also the whole testing process, particularly the pre- and post-examination procedures. In clinical practice, a discrepancy between genotype and phenotype may probably be caused by genotyping error. In this case, physicians may ask the laboratory to retest the patient sample or test the sample by an alternative method. Besides, the quality problems of genetic testing we summarized here are some common issues, which may be beneficial to medical department directors who want to conduct pharmacogenetic tests in the future. We did not assess the interpretation of genotyping results in this study. An important aspect of clinical pharmacogenetic testing is reporting the results. Future PT surveys will require participating laboratories to submit the predicted metabolizer phenotype along with genotype. In conclusion, the results of the first Chinese CYP2C9 and VKORC1 PT suggest that several laboratories were making mistakes, and the genotype information obtained with an in-house test should be interpreted with great care. Continuous external assessment and education in this field are needed, which will ensure that the referring clinician has the correct information for genotype-guided warfarin dosing. We recommend that all pharmacogenomic testing laboratories would regularly participate in PT schemes. REFERENCES 1. Johnson JA, Gong L, Whirl-Carrillo M, et al. Clinical Pharmacogenetics Implementation Consortium Guidelines for CYP2C9 and VKORC1 genotypes and warfarin dosing. Clin Pharmacol Ther. 2011;90:625–629. 2. Wadelius M, Pirmohamed M. Pharmacogenetics of warfarin: current status and future challenges. Pharmacogenomics J. 2007;7:99–111. 3. Fihn SD, McDonell M, Martin D, et al. Risk factors for complications of chronic anticoagulation. Ann Intern Med. 1993;118:511–520. 4. Budnitz DS, Pollock DA, Weidenbach KN, et al. National surveillance of emergency department visits for outpatient adverse drug events. JAMA. 2006;296:1858–1866. 5. Jorgensen AL, FitzGerald RJ, Oyee J, et al. Influence of CYP2C9 and VKORC1 on patient response to warfarin: a systematic review and metaanalysis. PLoS One. 2012;7:e44064. 6. Yang J, Chen Y, Li X, et al. Influence of CYP2C9 and VKORC1 genotypes on the risk of hemorrhagic complications in warfarin-treated patients: a systematic review and meta-analysis. Int J Cardiol. 2013;168: 4234–4243. 7. Sconce EA, Khan TI, Wynne HA, et al. The impact of CYP2C9 and VKORC1 genetic polymorphism and patient characteristics upon warfarin dose requirements: proposal for a new dosing regimen. Blood. 2005; 106:2329–2333. 8. Rieder MJ, Reiner AP, Gage BF, et al. Effect of VKORC1 haplotypes on transcriptional regulation and warfarin dose. N Engl J Med. 2005;352: 2285–2293.

134

| www.jcvp.org

9. Gage BF, Lesko LJ. Pharmacogenetics of warfarin: regulatory, scientific, and clinical issues. J Thromb Thrombolysis. 2008;25:45–51. 10. Pirmohamed M, Burnside G, Eriksson N, et al. A randomized trial of genotype-guided dosing of warfarin. N Engl J Med. 2013;369:2294– 2303. 11. Kimmel SE, French B, Kasner SE, et al. A pharmacogenetic versus a clinical algorithm for warfarin dosing. N Engl J Med. 2013;369: 2283–2293. 12. Do EJ, Lenzini P, Eby CS, et al. Genetics informatics trial (GIFT) of warfarin to prevent deep vein thrombosis (DVT): rationale and study design. Pharmacogenomics J. 2012;12:417–424. 13. Rakicevic LB, Kusic-Tisma JS, Kovac MK, et al. Rationalized DNA sequencing-based protocol for genotyping patients receiving coumarin therapy. Scand J Clin Lab Invest. 2013;73:523–527. 14. Okada Y, Nakamura K, Adachi A, et al. Development of a single-tube PCR-pyrosequencing method for the simultaneous and rapid detection of four variant alleles of CYP2C9 gene polymorphism. J Clin Pharm Ther. 2008;33:187–192. 15. Aomori T, Yamamoto K, Oguchi-Katayama A, et al. Rapid singlenucleotide polymorphism detection of cytochrome P450 (CYP2C9) and vitamin K epoxide reductase (VKORC1) genes for the warfarin dose adjustment by the SMart-amplification process version 2. Clin Chem. 2009;55:804–812. 16. Huang SW, Li Q, Zhu SY, et al. SYBR Green-based real-time PCR assay for detection of VKORC1 and CYP2C9 polymorphisms that modulate warfarin dose requirement. Clin Chem Lab Med. 2009;47:26–31. 17. King CR, Porche-Sorbet RM, Gage BF, et al. Performance of commercial platforms for rapid genotyping of polymorphisms affecting warfarin dose. Am J Clin Pathol. 2008;129:876–883. 18. Wu AH. Genotype and phenotype concordance for pharmacogenetic tests through proficiency survey testing. Arch Pathol Lab Med. 2013;137: 1232–1236. 19. Pratt VM, Zehnbauer B, Wilson JA, et al. Characterization of 107 genomic DNA reference materials for CYP2D6, CYP2C19, CYP2C9, VKORC1 and UGT1A1: a GeT-RM and Association for Molecular Pathology collaborative project. J Mol Diagn. 2010;12:835–846. 20. Higashi MK, Veenstra DL, Kondo LM, et al. Association between CYP2C9 genetic variants and anticoagulation-related outcomes during warfarin therapy. JAMA. 2002;287:1690–1698. 21. Tafe LJ, Datto MB, Palomaki GE, et al. Resource Committee. Molecular testing for the BRCA1 and BRCA2 Ashkenazi Jewish founder mutations: a report on the College of American Pathologists proficiency testing surveys. Genet Med. 2014;17:58–62. 22. Verhoef TI, Redekop WK, Daly AK, et al. Pharmacogenetic-guided dosing of coumarin anticoagulants: algorithms for warfarin, acenocoumarol and phenprocoumon. Br J Clin Pharmacol. 2014;77:626–641. 23. Miao L, Yang J, Huang C, et al. Contribution of age, body weight, and CYP2C9 and VKORC1 genotype to the anticoagulant response to warfarin: proposal for a new dosing regimen in Chinese patients. Eur J Clin Pharmacol. 2007;63:1135–1141. 24. Zhong SL, Yu XY, Liu Y, et al. Integrating interacting drugs and genetic variations to improve the predictability of warfarin maintenance dose in Chinese patients. Pharmacogenet Genomics. 2012;22:176–182. 25. Lou Y, Han L, Li Y, et al. Impact of six genetic polymorphisms on Warfarin maintenance dose variation in Chinese Han population [in Chinese]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi. 2014;31:367–371. 26. Zuo J, Xia D, Jia L, et al. Genetic polymorphisms of drug-metabolizing phase I enzymes CYP3A4, CYP2C9, CYP2C19 and CYP2D6 in Han, Uighur, Hui and Mongolian Chinese populations. Pharmazie. 2012;67: 639–644. 27. Langley MR, Booker JK, Evans JP, et al. Validation of clinical testing for warfarin sensitivity: comparison of CYP2C9-VKORC1 genotyping assays and warfarin-dosing algorithms. J Mol Diagn. 2009;11: 216–225. 28. Chen C, Li S, Lu X, et al. High resolution melting method to detect single nucleotide polymorphism of VKORC1 and CYP2C9. Int J Clin Exp Pathol. 2014;7:2558–2564. 29. Yang S, Xu L, Wu HM. Rapid genotyping of single nucleotide polymorphisms influencing warfarin drug response by surface-enhanced laser desorption and ionization time-of-flight (SELDI-TOF) mass spectrometry. J Mol Diagn. 2010;12:162–168.



A PCR method for VKORC1 G-1639A and CYP2C9 A1075C genotyping useful to warfarin therapy among Japanese.

Identification of CYP2C9 and VKORC1 polymorphisms in Iranian patients who are under warfarin therapy.

Polymorphisms of VKORC1 and CYP2C9 are associated with warfarin sensitivity in Chinese population.

The effect of CYP2C9 and VKORC1 genetic polymorphisms on warfarin dose requirements in a pediatric population.

VKORC1 and CYP2C9 Genotype Variations in Relation to Warfarin Dosing in Korean Stroke Patients.

Warfarin Dosing in a Patient with CYP2C9(∗)3(∗)3 and VKORC1-1639 AA Genotypes.

Variability of warfarin dose response associated with CYP2C9 and VKORC1 gene polymorphisms in Chinese patients.

The influence of VKORC1 and CYP2C9 mutations on warfarin response after total hip and knee arthroplasty.

VKORC1 Genotype and Anticoagulation Control Quality.

Clinical application of a new warfarin-dosing regimen based on the CYP2C9 and VKORC1 genotypes in atrial fibrillation patients.

High resolution melting method to detect single nucleotide polymorphism of VKORC1 and CYP2C9.

Genetic epidemiology of pharmacogenetic variations in CYP2C9, CYP4F2 and VKORC1 genes associated with warfarin dosage in the Indian population.

Genetic polymorphisms in warfarin and tacrolimus-related genes VKORC1, CYP2C9 and CYP3A5 in the Greek-Cypriot population.

Warfarin dose requirement in Turkish patients: the influences of patient characteristics and polymorphisms in CYP2C9, VKORC1 and factor VII.

Prevalence of genetic polymorphisms of CYP2C9 and VKORC1 - implications for warfarin management and outcome in Croatian patients with acute stroke.

The impact of age and CYP2C9 and VKORC1 variants on stable warfarin dose in the paediatric population.

National Prociency Testing Result of CYP2D6*10 Genotyping for Adjuvant Tamoxifen Therapy in China.

Variation in genes controlling warfarin disposition and response in American Indian and Alaska Native people: CYP2C9, VKORC1, CYP4F2, CYP4F11, GGCX.

Impact of CYP2C9 and VKORC1 genetic polymorphisms upon warfarin dose requirements in Egyptian patients with acute coronary syndrome.

Implementing Algorithm-Guided Warfarin Dosing in an Ethnically Diverse Patient Population Using Electronic Health Records and Preemptive CYP2C9 and VKORC1 Genetic Testing.

Quality score compression improves genotyping accuracy.

Novel single nucleotide polymorphism in CYP2C9 is associated with changes in warfarin clearance and CYP2C9 expression levels in African Americans.

Mechanistic insights into the effect of CYP2C92 and CYP2C93 variants on the 7-hydroxylation of warfarin.

Genetic polymorphisms of CYP2C9, CYP2C19, and CYP3A5 in Kosovar population.