Evaluation of a Microarray-Based Genotyping Assay for the Rapid ...

Clinical Chemistry / Microarray-Based CYP2C19 Genotyping Assay

Evaluation of a Microarray-Based Genotyping Assay for the Rapid Detection of Cytochrome P450 2C19 *2 and *3 Polymorphisms From Whole Blood Using Nanoparticle Probes Blake W. Buchan, PhD,1,2 Jess F. Peterson, MD,1 Christopher H. Cogbill, MD,1 Dennis K. Anderson,3 Joellen S. Ledford,3 Mary N. White,3 Neil B. Quigley, PhD,3 Paul J. Jannetto, PhD,1,2 and Nathan A. Ledeboer, PhD, D(ABMM)1,2 Key Words: CYP2C19; Polymorphism; Molecular detection assay; Nanoparticle DOI: 10.1309/AJCPCPU9Q2IRNYXC

Abstract Numerous drugs such as clopidogrel have been developed to reduce coagulation or inhibit platelet function. The hepatic cytochrome P450 (CYP) pathway is involved in the conversion of clopidogrel to its active metabolite. A recent black-box warning was included in the clopidogrel package insert indicating a significant clinical link between specific CYP2C19 genetic variants and poor metabolism of clopidogrel. Of these variants, *2 and *3 are the most common and are associated with complete loss of enzyme activity. In patients who are carriers of a CYP2C19 *2 or *3 allele, the conversion of clopidogrel to its active metabolite may be reduced, which can lead to ischemic events and negative consequence for the patient. We examined the ability of the Verigene CLO assay (Nanosphere, Northbrook, IL) to identify CYP2C19 *2 and *3 polymorphisms in 1,286 unique whole blood samples. The Verigene CLO assay accurately identified homozygous and heterozygous *2 and *3 phenotypes with a specificity of 100% and a final call rate of 99.7%. The assay is fully automated and can produce a result in approximately 3.5 hours.

604 604

Am J Clin Pathol 2011;136:604-608 DOI: 10.1309/AJCPCPU9Q2IRNYXC

Thrombotic ischemic events resulting from atrial fibrillation, atherosclerosis, or hypercoagulation disorders or following coronary artery surgery can cause severe, lifethreatening conditions including stroke, acute myocardial infarction, and pulmonary embolism.1-3 There are a number of US Food and Drug Administration–approved drugs recommended for use by the American College of Cardiology and the American Heart Association that have been evaluated for anticoagulation or antiplatelet therapy.4 The effectiveness and safety of these drugs depends largely on appropriate dosing. An incorrect treatment regimen can result in recurrent ischemic events or severe life-threatening hemorrhage. The appropriate dose of a given drug for a person is determined by the pharmacokinetic properties of the drug, which can be affected by various factors including absorption, distribution, metabolism, and excretion rates. The CYP proteins are a large group of enzymes found in high concentration in the liver and are involved in the metabolism of up to 90% of prescribed drugs.5,6 The CYP enzymes represent at least 50 families and several subfamilies, each encoded by a number of individual genes.6 Specifically, CYP2C19 is involved in the metabolism of up to 25% of prescribed drugs, including proton pump inhibitors, antidepressants, antiepileptics, and anticoagulants such as clopidogrel.5,7 CYP2C19 has an integral role in the activation of clopidogrel by aiding in its conversion from an inactive prodrug to an active metabolite capable of inhibiting platelet function.7-9 The wild-type CYP2C19 allele, *1, imparts fully functional metabolism of clopidogrel. Numerous polymorphisms in CYP2C19, including *2, *3, *4, *5, *6, *7, *8, and *17 have been identified that can result in increased or decreased activity of the enzyme.10 The 2 most prevalent © American Society for Clinical Pathology

Clinical Chemistry / Original Article

variant alleles are G>A transitions denoted as *2 (681G>A) and *3 (636G>A).10-12 The frequency of the *2 allele is 14% to 17% among populations of African or European descent and 30% within populations of Chinese descent. The *3 allele is the second most frequent polymorphism, present at frequencies of 0.04%, 0.4%, and 5.1% in European, African, and Chinese descent populations, respectively.13 The *2 and *3 polymorphisms result in a complete loss of enzyme activity due to truncation of the enzyme and cause a “poor metabolizer” phenotype.11,12,14 A number of studies have suggested that decreased activity of CYP2C19 due to the *2 polymorphism contributes to reduced efficacy of clopidogrel to prevent recurrent ischemic events leading to myocardial infarction, stent thrombosis, and ischemic stroke.9,15-17 The impact of alternative clopidogrel dosing in patients with these polymorphisms has been examined, and a second or third loading dose was demonstrated to reduce platelet reactivity during therapy.16 Recently, the clopidogrel package insert was modified to include a black-box warning indicating a significant clinical link between CYP2C19 lossof-function genotypes (*2 and *3) and poor metabolism of clopidogrel. Therefore, a rapid method to identify these polymorphisms could aid in guiding clopidogrel therapy or allow selection of an alternative antiplatelet drug. In this study, we assessed the performance of the investigational-use-only (IUO) Verigene Clopidogrel Metabolism Nucleic Acid Test (Verigene CLO; Nanosphere, Northbrook, IL) to identify CYP2C19 *2 and *3 polymorphisms from whole blood. The Verigene CLO IUO assay is an automated sample-to-result microarray-based assay in which DNA is extracted from whole blood samples and hybridized to allele-specific probes immobilized on a glass slide. Detection of captured DNA is achieved using nanoparticleconjugated probes that have been demonstrated to provide excellent sensitivity and that eliminate the need for a target amplification step before array hybridization.18,19

Materials and Methods Blood specimens used in the study were obtained from LifeSource (Chicago, IL) or Blood Centers of the Pacific (San Francisco, CA) and were deidentified before analysis in accordance with the institutional review board–approved protocol at each site. Blood specimens were kept at 4ºC for a maximum of 8 days before analysis. Specimens were analyzed using the Verigene CLO IUO assay at 1 of 3 sites: Nanosphere, 185 specimens; the Medical College of Wisconsin, Milwaukee, 561 specimens; and the Molecular Pathology Laboratory Network, Maryville, TN, 542 specimens. Vials of blood were homogenized by rotation for 30 minutes before beginning the assay. To set up the assay, a

single-use processing tray containing all necessary reagents to lyse, extract, and purify DNA from whole blood specimens was loaded into the Verigene Processor SP (Nanosphere). A 1.0-mL sample of homogenized blood was transferred to the specimen well in the extraction tray. A single-use CLO test cartridge containing the slide array and hybridization reagents was loaded into the Verigene Processor SP, and the assay was started (run time, 3 hours, 8 minutes). On completion of the assay, the CLO test cartridge was removed from the processor and the hybridization slide was inserted into the Verigene Reader. The reader returned results of *1/*1, *1/*2, *2/*2, *1/*3, *3/*3, or *2/*3 within approximately 60 seconds. A message of “no call” was returned if the genotype of the specimen could not be definitively identified owing to specimen deficiency or assay error. Testing of these specimens was repeated, provided the original specimen was sufficient for retesting, ie, 1 mL or more remained at 8 days or less after collection. Residual DNA from each specimen was frozen and sent to ACGT (Wheeling, IL) for sequencing to confirm the CYP2C19 genotype. Briefly, target regions of the DNA [CYP2C19 *2 (rs4244285) G681A and CYP2C19 *3 (rs57081121) G636A] were amplified from the DNA samples using proprietary oligonucleotide primers. Sequencing of each strand was conducted using the appropriate oligonucleotide primer from the initial amplification reaction using BigDye terminator chemistry (V 3.1; Applied Biosystems, Carlsbad, CA). The sequencing primer extension reactions were analyzed on an ABI 3730XL or 3730 DNA Analyzer (Applied Biosystems). Top and bottom strand sequences were aligned, and the assembled data were compared with the reference sequence. Specimens used for the reproducibility study conformed to the same testing guidelines as those used for the clinical trial, ie, deidentified and stored at 4ºC for a maximum of 8 days before analysis. A single blood specimen was obtained and divided among the 3 test sites for reproducibility analysis. Reproducibility testing for each specimen was conducted a minimum of twice daily by 2 different operators on 5 different days.

Results A total of 1,286 unique blood specimens were analyzed at 3 test sites using the Verigene CLO assay to identify CYP2C19 *2 and *3 polymorphism phenotypes. Bidirectional sequencing conducted on 1,264 of these specimens indicated the presence of heterozygous and homozygous *2 and *3 polymorphisms within this group. Specifically, the study cohort included 436 *1/*2 heterozygous, 89 *2/*2 homozygous, 56 *1/*3 heterozygous, 7 *3/*3 homozygous, and 29 *2/*3 heterozygous specimens ❚Table 1❚. The remaining 647 specimens were

© American Society for Clinical Pathology

Am J Clin Pathol 2011;136:604-608 605

DOI: 10.1309/AJCPCPU9Q2IRNYXC

605 605

Buchan et al / Microarray-Based CYP2C19 Genotyping Assay

❚Table 1❚ Study Enrollment Characteristics and Ability of the Verigene CLO Assay to Generate Definitive Genotype Results† Genotype‡

Specimens

Initial Call

Retest Call

Final Call

*1/*1§

647 (50.3) 436 (33.9) 89 (6.9) 56 (4.4) 7 (0.5) 29 (2.3) 22 (1.7) 1,286

613/647 (94.7) 416/436 (95.4) 85/89 (96) 53/56 (95) 6/7 (86) 29/29 (100) 0/22 (0) 1,202/1,286 (93.5)

33/34 (97) 17/20 (85) 4/4 (100) 3/3 (100) 1/1 (100) — — 58/62 (94)

646/647 (99.8) 433/436 (99.3) 89/89 (100) 56/56 (100) 7/7 (100) 29/29 (100) — 1,260/1,264 (99.7)

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

Unknown|| Total †

Data are given as number (percentage) or number/total (percentage). Genotype based on bidirectional sequencing results. genotype inferred; negative for *2 and *3. || Specimens that could not be reanalyzed within experimental guidelines, undetermined genotype. ‡

§ *1/*1

negative for *2 and *3 polymorphisms and were inferred to be *1/*1 (wild-type) homozygous (Table 1). An initial result was obtained for 1,202 of 1,286 specimens with the Verigene CLO test, resulting in an initial call rate of 93.5% (Table 1). Of the 84 specimens that did not give a definitive result, 62 could be repeated within the experimental guidelines. On retesting, 58 (94%) of 62 generated a definitive result. Excluding the 22 specimens that could not be reanalyzed, the final call rate was 99.7% (Table 1). A comparison of bidirectional sequencing results with results generated using the Verigene CLO assay revealed 100% final concordance for all 1,260 specimens that were successfully analyzed ❚Table 2❚.

The 84 specimens that failed to generate a definitive result on initial analysis could be grouped into 3 categories based on cause of failure: “No Call, Variation,” “No Call, Background,” and “No Call, No Grid” ❚Table 3❚. No Call, Variation is the result of failure of the specimen to meet detection algorithm criteria, while No Call, Background and No Call, No Grid reflect mechanical failures of the test cartridges owing to high background signal and failure to detect the microarray grid, respectively. Analysis failure did not seem to correlate with any specific genotype, and reanalysis of the specimens with an initial no call resulted in a 93.5% call rate (final call rate, initial + reanalysis is 99.7%).

❚Table 2❚ Final Accuracy of the Verigene CLO Assay for Identifying CYP2C19 Polymorphisms Bidirectional Sequencing Result Verigene CLO Result

*1/*1

*1/*2

*2/*2

*1/*3

*3/*3

*2/*3

*1/*1

646† 0 0 0 0 0

0 433 0 0 0 0

0 0 89 0 0 0

0 0 0 56 0 0

0 0 0 0 7 0

0 0 0 0 0 29

*1/*2 *2/*2 *1/*3 *3/*3 *2/*3 † *1/*1

genotype inferred; negative for *2 and *3.

❚Table 3❚ Source of Analysis Failure for Specimens Not Generating Definitive Results Failure Type

All Genotypes

*1/*1

*1/*2

*2/*2

*1/*3

*3/*3

*2/*3

Definitive Retest†

Not Retested‡

Variation Background No Grid Total

54 23 7 84

27 4 2 33

12 3 2 17

0 2 2 4

3 0 0 3

1 0 0 1

0 0 0 0

43§ 9|| 6 58

9 12 1 22

†

Number of specimens that failed initial analysis but gave definitive result on retesting. Number of specimens that could not be reanalyzed within experimental guidelines. § The 2 specimens that did not give definitive results on retesting were “No Call, Variation,” or failure of the specimen to meet detection algorithm criteria. || The 2 specimens that did not give definitive results on retesting were 1 each “No Call, Variation” and “No Call, Background.” No Call, Background reflects mechanical failure of the test cartridges owing to high background signal. ‡

606 606



Clinical Chemistry / Original Article

We conducted a reproducibility study to examine the variability of the Verigene CLO to accurately identify CYP2C19 *2 and *3 polymorphisms within clinical specimens and between different test sites and operators. A single clinical specimen of each of the following genotypes, *1/*1, *1/*2, *2/*2, *1/*3, and *2/*3, was tested at least twice daily on 5 nonconsecutive days at 3 different sites for a minimum of 30 potential results. The call rate ranged from 90% to 100% depending on the specimen and was 96.1% overall ❚Table 4❚. The accuracy of Verigene CLO results during the reproducibility study was 100%. These call rates and accuracy results are similar to those observed during the course of the methods comparison study. The total time to result was approximately 3.5 hours, which included a 30-minute blood homogenization by rocking before loading specimens on the Verigene instrument.

Discussion Current techniques to identify people not responding to clopidogrel antiplatelet therapy include light transmission aggregometry and flow cytometry–based vasodilatorstimulated phosphoprotein assays.20 A drawback to these methods is that they are restricted to analysis of postdose blood specimens. Therefore, people with a poor metabolizer phenotype are not identified until after therapy has begun, which can delay appropriate dosing or drug choice. There is 1 currently available genotyping method, the AmpliChip CYP450 Test (Roche, Basel, Switzerland) that is cleared by the Food and Drug Administration for use in the United States. It is capable of identifying CYP2C19 *2 and *3 polymorphisms and several CYP2D6 polymorphisms. One study found 95.6% concordance between the AmpliChip CYP450 Test and polymerase chain reaction–restriction fragment length polymorphism analysis for identification of CYP2D6 polymorphisms; however, no study has examined the specificity of the AmpliChip CYP450 Test to identify CYP2C19 polymorphisms.21 The AmpliChip CYP450 Test

requires manual extraction and purification of DNA from blood specimens and a target amplification step before hybridization to the array. These additional steps can introduce the possibility for error or variation during the manual extraction and amplification steps and extend the total assay time to up to 24 hours. The Verigene CLO assay identified heterozygous and homozygous *2 and *3 polymorphisms with 100% concordance to bidirectional sequence analysis from 1,260 unique clinical blood specimens. The initial and final call rates were 93.5% and 99.7%, respectively. Highly reproducible results and call rates among different test sites and operators suggest low variability based on technologist or assay setup. Of the 62 specimens that failed to generate a definitive result on initial analysis, only 4 failed to generate a definitive result on reanalysis. A possible explanation for consecutive failed analyses would be an insufficient DNA concentration in the specimen. This could be the result of poor DNA extraction on 2 independent tests; however, this seems unlikely. Repeated failed analysis could also be the result of a specimen with an overall low DNA concentration, such as that obtained from a leukocytopenic donor. A major advantage of the Verigene CLO assay is the fully automated sample-to-result capability. This feature decreases the possibility for processing error and reduces the total assay time to approximately 3.5 hours with little “hands-on” time for technologists. The use of gold nanoparticle-conjugated probes to detect nucleic acids captured by array probes greatly enhances the detection sensitivity and eliminates the need for target amplification before array hybridization.18 This also eliminates the chance of false priming and amplification of nonspecific target DNA. In addition, the Verigene Processor SP is a small bench-top instrument that can be placed within hospital or clinic laboratories for on-demand testing, which eliminates the need for send-out testing and further reduces the time to result. A limitation of the Verigene CLO assay, as well as of the AmpliChip CYP450 Test and other genotyping methods,

❚Table 4❚ Reproducibility Characteristics Across Three Study Sites† Definitive Results (Call Rate) Specimen Genotype

Site A

Site B

Site C

Total

*1/*1‡

10/10 (100) 10/10 (100) 10/10 (100) 10/10 (100) 10/10 (100) 50/50 (100)

10/10 (100) 7/10 (70) 10/10 (100) 11/11 (100) 10/10 (100) 48/51 (94)

11/11 (100) 10/10 (100) 10/10 (100) 9/10 (90) 9/11 (82) 49/52 (94)

31/31 (100) 27/30 (90) 30/30 (100) 30/31 (97) 29/31 (94) 147/153 (96.1)

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

Total †

Data are given as number/total (percentage). genotype inferred; negative for *2 and *3.

‡ *1/*1


Am J Clin Pathol 2011;136:604-608 607

DOI: 10.1309/AJCPCPU9Q2IRNYXC

607 607

Buchan et al / Microarray-Based CYP2C19 Genotyping Assay

is the incomplete knowledge of all genetic and physiologic factors contributing to altered metabolism of clopidogrel and other drugs. Current assays focus on well-defined CYP2C19 *2 and *3 polymorphisms; however, other enzymes in the CYP450 family, including 2B6, 2C9, and 3A4, may also have redundant roles in the metabolism of clopidogrel.7 In addition, other clinical factors, including poor absorption, diabetes mellitus, and elevated body mass index, have been proposed to influence individual response to clopidogrel therapy.8 Inexpensive genotyping assays that would be available for use in hospital laboratories and clinics have the potential to aid in guiding anticoagulation therapy and prevent potential life-threatening complications. The Verigene CLO assay identifies CYP 450 2C19 *2 and *3 polymorphisms associated with diminished response to clopidogrel with 100% specificity in as little as 3.5 hours. Further studies will need to be conducted to assess the clinical usefulness of this and similar genetic tests to dictate therapeutic course and improve patient outcome. From the 1Department of Pathology, Medical College of Wisconsin and 2Dynacare Laboratories, Milwaukee, WI; and 3Molecular Pathology Laboratory Network, Maryville, TN. Test materials and financial support for this study were provided by Nanosphere, Northbrook, IL. Address reprint requests to Dr Ledeboer: Dept of Pathology, Medical College of Wisconsin, 9200 W Wisconsin Ave, Milwaukee, WI 53226. Dr Ledeboer is a consultant to Nanosphere.

References 1. Kujovich JL. Factor V Leiden thrombophilia. Genet Med. 2011;13:1-16. 2. Schwann TA, Kistler L, Engoren MC, et al. Incidence and predictors of postoperative deep vein thrombosis in cardiac surgery in the era of aggressive thromboprophylaxis. Ann Thorac Surg. 2010;90:760-766. 3. Albers GW, Amarenco P, Easton JD, et al. Antithrombotic and thrombolytic therapy for ischemic stroke: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest. 2008;133(6 suppl):630S669S. 4. Bonaca MP, Steg PG, Feldman LJ, et al. Antithrombotics in acute coronary syndromes. J Am Coll Cardiol. 2009;54:969984. 5. Li X, Quigg RJ, Zhou J, et al. Clinical utility of microarrays: current status, existing challenges and future outlook. Curr Genomics. 2008;9:466-474. 6. Lynch T, Price A. The effect of cytochrome P450 metabolism on drug response, interactions, and adverse effects. Am Fam Physician. 2007;76:391-396.

608 608


7. Kazui M, Nishiya Y, Ishizuka T, et al. Identification of the human cytochrome P450 enzymes involved in the two oxidative steps in the bioactivation of clopidogrel to its pharmacologically active metabolite. Drug Metab Dispos. 2010;38:92-99. 8. Angiolillo DJ, Fernandez-Ortiz A, Bernardo E, et al. Variability in individual responsiveness to clopidogrel: clinical implications, management, and future perspectives. J Am Coll Cardiol. 2007;49:1505-1516. 9. Hulot JS, Collet JP, Silvain J, et al. Cardiovascular risk in clopidogrel-treated patients according to cytochrome P450 2C19*2 loss-of-function allele or proton pump inhibitor coadministration: a systematic meta-analysis. J Am Coll Cardiol. 2010;56:134-143. 10. Home Page of the Human Cytochrome P450 (CYP) Allele Nomenclature Committee. http://www.cypalleles.ki.se. Last updated September 9, 2008. Accessed March 2010. 11. de Morais SM, Wilkinson GR, Blaisdell J, et al. Identification of a new genetic defect responsible for the polymorphism of (S)-mephenytoin metabolism in Japanese. Mol Pharmacol. 1994;46:594-598. 12. de Morais SM, Wilkinson GR, Blaisdell J, et al. The major genetic defect responsible for the polymorphism of S-mephenytoin metabolism in humans. J Biol Chem. 1994;269:15419-15422. 13. Xie HG, Kim RB, Wood AJ, et al. Molecular basis of ethnic differences in drug disposition and response. Annu Rev Pharmacol Toxicol. 2001;41:815-850. 14. Gurbel PA, Tantry US, Shuldiner AR, et al. Genotyping: one piece of the puzzle to personalize antiplatelet therapy. J Am Coll Cardiol. 2010;56:112-116. 15. Jin B, Ni H-C, Shen W, et al. Cytochrome P450 2C19 polymorphism is associated with poor clinical outcomes in coronary artery disease patients treated with clopidogrel [published online ahead of print September 16, 2010]. Mol Biol Rep. 2011;38:1697-1702. doi:10.1007/s11033-010-0282-0. 16. Bonello L, Armero S, Ait Mokhtar O, et al. Clopidogrel loading dose adjustment according to platelet reactivity monitoring in patients carrying the 2C19*2 loss of function polymorphism. J Am Coll Cardiol. 2010;56:1630-1636. 17. Shuldiner AR, O’Connell JR, Bliden KP, et al. Association of cytochrome P450 2C19 genotype with the antiplatelet effect and clinical efficacy of clopidogrel therapy. JAMA. 2009;302:849-857. 18. Storhoff JJ, Marla SS, Bao P, et al. Gold nanoparticlebased detection of genomic DNA targets on microarrays using a novel optical detection system. Biosens Bioelectron. 2004;19:875-883. 19. Jannetto PJ, Buchan BW, Vaughan KA, et al. Real-time detection of influenza A, influenza B, and respiratory syncytial virus A and B in respiratory specimens by use of nanoparticle probes. J Clin Microbiol. 2010;48:3997-4002. 20. Calatzis A, Leitner M, Panzer S. Monitoring anticoagulation of primary haemostasis: estimation of platelet function in whole blood assays. Hamostaseologie. 2009;29:279-284. 21. Heller T, Kirchheiner J, Armstrong VW, et al. AmpliChip CYP450 GeneChip: a new gene chip that allows rapid and accurate CYP2D6 genotyping. Ther Drug Monit. 2006;28:673677.
