Comparison of performance of three commercial platforms for warfarin sensitivity genotyping Nikolina Babic a , Eden V. Haverfield b , Julie A. Burrus a , Anthony Lozada b , Soma Das b , Kiang-Teck J. Yeo a, ⁎ a Department of Pathology, Pritzker School of Medicine, The University of Chicago, 5481 S. Maryland Avenue, Chicago, IL 60637, United States b Department of Human Genetics, Pritzker School of Medicine, The University of Chicago, 5481 S. Maryland Avenue, Chicago, IL 60637, United States abstract article info Article history: Received 24 April 2009 Received in revised form 9 June 2009 Accepted 10 June 2009 Available online 21 June 2009 Keywords: Warfarin Genotyping Pharmacogenetics Cytochrome P-450 CYP2C9 VKORC1 Background: We performed a 3-way comparison on the Osmetech eSensor®, AutoGenomics INFINITI™, and a real-time PCR method (Paragonx™ reagents/Stratagene® RT-PCR platform) for their FDA-cleared warfarin panels, and additional polymorphisms (CYP2C9 ⁎5, ⁎6, and ⁎11 and extended VKORC1 panels) where available. Methods: One hundred de-identified DNA samples were used in this IRB-approved study. Accuracy was deter- mined by comparison of genotyping results across three platforms. Any discrepancy was resolved by bi- directional sequencing. The CYP4F2 on Osmetech was validated by bi-directional sequencing. Results: Accuracies for CYP2C9 ⁎2 and ⁎3 were 100% for all 3 platforms. VKORC1 3673 genotyping accuracies were 100% on eSensor and 97% on Infiniti. CYP2C9 ⁎5, ⁎6 and ⁎11 showed 100% concordance between eSensor and Infiniti. VKORC1 6484 and 9041 variants compared between ParagonDx and Infiniti analyzer were 100% (6484) and 99% (9041) concordant. CYP4F2 was 100% concordant with sequencing results. The time required to generate the results from automated DNA extraction-to-result was approximately 8 h on Infiniti, and 4 h on eSensor and ParagonDx, respectively. Conclusions: Overall, we observed excellent CYP2C9 ⁎2 and ⁎3 genotyping accuracy for all three platforms. For VKORC1 3673 genotyping, eSensor demonstrated a slightly higher accuracy than the Infiniti, and CYP4F2 on Osmetech was 100% accurate. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Warfarin is a widely prescribed anticoagulant drug, with up to a million new patients initiated on therapy each year for prevention of thromboembolic events due to atrial fibrillation, heart valves, and orthopedic surgery [1,2]. The major challenge in warfarin administra- tion is selecting a correct dose schedule for a given patient due to the drug's narrow therapeutic index and large inter-individual variability in drug disposition and effect. Consequently, patients who receive war- farin doses higher than necessary are at a significant risk of develop- ing life-threatening hemorrhage, while sub-therapeutic dose might lead to thrombosis. Warfarin dosing is further complicated by its inter- individual dose variability, depending on factors such as diet, age, body size, smoking status, race, concomitant diseases, drug interactions and genetic variability [3,4]. Warfarin therapy is monitored using Inter- national Normalized Ratio (INR), which is reported as the ratio of the patient's prothrombin time to that of a reference population. Based on several anticoagulation trials, the target INR levels are between 2 and 3 [5,6]. Currently, algorithms based on clinical factors explain approxi- mately 20% of the variance in warfarin dosing and do not include pharmacogenomics parameters [7]. Numerous studies have shown that polymorphisms in cytochrome P-450 2C9 (CYP2C9) and vitamin K epoxide reductase complex (VKORC1) genes play significant roles in determination of warfarin sensitivity and resistance [8–14]. Patients with CYP2C9 ⁎2 (c.430CNT, [p.R144C], or rs1799853) and ⁎3 (c.1075ANG, [p.I359L], or rs1057910) alleles require significantly lower maintenance doses, longer times to stable INR and are at a higher risk of bleeding. Using CYP2C9 genotyping only, Caraco et al. found that the average time outside the target INR was lower for the genotype-guided group than controls [8]. In addition to warfarin sensitivity, there are rare cases of warfarin resistance. This concept has come to light only recently with the discovery of VKORC1 gene. Rieder et al. identified 2 major VKORC1 haplotype groups that were respon- sible for changes in warfarin sensitivity: a low-dose haplotype group (A) and a high-dose haplotype group (B) [15]. Haplotype group A in- cludes VKORC1 SNP 3673 that has been associated with increased war- farin sensitivity. In a study of 297 patients starting on warfarin therapy, Schwarz et al. found that the patients with A/A haplotype of VKORC1 had a decreased time to the first therapeutic INR [12]. Pharmacogenomic models for warfarin dosing are based on large prospective, randomized clinical trials [1,9,11]. For instance, Gage et al. developed a drug dosing model that can account for about 60% of the Clinica Chimica Acta 406 (2009) 143–147 Abbreviations: ASPE, allele-specific primer extension; ASR, analyte specific reagents; LD, linkage disequilibrium; VKORC1 , vitamin K epoxide reductase complex 1. ⁎ Corresponding author. Tel.: +1773 702 1318; fax: +1 773 702 6268. E-mail address: jyeo@bsd.uchicago.edu (K.-T.J. Yeo). 0009-8981/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2009.06.015 Contents lists available at ScienceDirect Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim