2.1 Patients
We enrolled patients who were regularly administered warfarin (Orion, Espoo, Finland) anticoagulant therapy for more than 1 month in the First Affiliated Hospital of Fujian Medical University as a part of their atrial fibrillation treatment (paroxysmal atrial fibrillation, persistent atrial fibrillation, and permanent atrial fibrillation), deep vein thrombosis treatment, or during heart valve replacement. An international normalized ratio (INR) between 1.5 and 3.0 measured three times in succession was considered as the maintenance dose (Figure 1). The liver and kidney function indicators of all enrolled patients were normal. Demographic characteristics, genetic factors, and concomitant medication data were collected from the medical records of each enrolled patient in this study (Table 1).
The exclusion criteria were as follows: a diagnosis of hematological or coagulopathy disorder; heart function grade III or above; clear hemorrhagic disease or bleeding tendency; liver and kidney dysfunction; severe malnutrition; pregnancy; long-term administration of drugs that might interfere with warfarin function and metabolism (such as aspirin, clopidogrel, heparin and vitamin K, amiodarone, rifampicin, and barbital); bleeding or thromboembolism during observation; and missing clinical and laboratory-related examination data.
2.2 Blood withdrawal and sample treatment
Two milliliters of blood was collected from each patient in the steady state at 12–15 h post warfarin administration into an EDTA anticoagulation tube. Whole blood (200 μL) was used for genotyping (see Section 2.5), and the remaining volume of the suspended plasma was stored at -80°C. The concentrations of R- and S-warfarin were measured by UPLC-MS/MS within 5 days of blood collection.
For each calibration curve standard, quality control sample, and patient sample, 300 μL of plasma was pipetted into a 2-mL test tube containing 10 μL of internal standard solution (carbamazepine; China National Institute for the Control of Pharmaceutical and Biological Products, Beijing, China) and 100 μL of HCl (1 M) solution. Proteins were denatured by adding 2 mL of dichloromethane and vortexed for 3 min, followed by centrifugation for 5 min at 2550 ×g. The organic phase was transferred into a vial and dried with nitrogen. The residue was dissolved in 200 μL of mobile phase (see Section 2.4) and filtered through a 0.22-μm microporous membrane. Ten microliters of this solution was injected into the UPLC-MS/MS system (see Section 2.4) for analysis.
2.3 Analysis
Chromatography was performed using the Waters Acquity UPLC system (Milford, MA, USA) with an Astec Chirobiotic V column (250 mm × 2.1 mm, 5 μm) (Supelco, USA) at 25℃. The mobile phase comprised 55% mobile phase A (0.02% formic acid in water) and 45% mobile phase B (methanol). The flow rate was 0.2 mL/min.
MS detection was performed on a Waters TQD tandem mass spectrometer equipped with an electrospray ionization source. The mass spectrometer was set at the multiple reaction monitoring mode and quantification was performed at m/z 307 >161.18 for warfarin (cone voltage: 50 V) and m/z 237.31 >194 for carbamazepine (cone voltage: 40 V). The collision energy was 20 V.
2.4 Validation procedures
Calibration standards were prepared by spiking blank plasma samples with standard solutions. The final concentrations of the calibration samples were 4.3, 8.6, 26.0, 65.0, 130.0, 260.0, 520.0, and 1040.0 µg/L for R-warfarin (Toronto Research Chemicals, North York, Canada) and 4.8, 9.6, 28.6, 71.5, 143.0, 286.0, 572.0, and 1144.0 µg/L for S-warfarin (Toronto Research Chemicals).
Accuracy and intra- and inter-day precisions were determined using pooled plasma (n = 5) at three different concentrations: lower limit of quantitation, high QC (approximately 80% upper limit of quantitation), and upper limit of quantitation. A signal-to-noise ratio of ≥3 was used as the lower limit of detection.
2.5 Genotyping
DNA was extracted from blood samples using the adsorption column method with the Blood Genomic DNA Extraction Kit (Beijing Tiangen Biotech, Beijing, China) within 24 h of blood collection. Genotyping was carried out by PCR-RFLP, and primers were designed by Shanghai Shenggong Bioengineering [5-9]. The restriction enzymes Sau3aI, SmaI, PvuII, NsiI, and MspI were provided by Promega (Madison, WI, USA). The results were verified by multiplex PCR and sequencing (HiSeq XTen sequencers, Illumina, San Diego, CA, USA). A panel containing 25 target SNP sites was designed. Library preparation was performed by two-step PCR. The plate was sealed and the PCR was carried out in a thermal cycler (T100TM, Bio-Rad, Hercules, CA, USA) using the following program: one cycle of denaturation at 95°C for 3 min, five cycles of denaturation at 94°C for 30 s, annealing at 55°C for 20 s, and elongation at 72°C for 30 s, with a final extension at 72°C for 5 min. AMPure XP beads were used to purify the amplicon product. The libraries were then quantified and pooled. Paired-end sequencing of the library was performed on the HiSeq XTen sequencer (Illumina).
2.6 Statistics
The probability of deviation from the Hardy–Weinberg equilibrium was calculated for each SNP. Skewness and kurtosis were used to test whether the blood concentration values were normally distributed (SPSS 17.0, IBM, Armonk, NY, USA). One-way analysis of variance was used to evaluate data showing a normal distribution, and Student–Newman–Keuls q-test was used for pairwise comparison between groups. Kruskal–Wallis test was used to compare the steady-state plasma concentrations in patients among different genotype groups. A P value of < 0.05 was considered to indicate statistical significance.