B-163 Development and Validation of a New Quantitative LC-MS/MS Method for Vitamin K1 in Plasma and Serum

Abstract Background Vitamin K1 (phylloquinone) is a fat-soluble cofactor; its deficiency depresses synthesis of plasma prothrombin (factor II) and other factors, causes excessive bleeding, and increases the risk of bone fractures and osteoporosis. Biochemical and metabolic studies on vitamin K1 require both highly sensitive and specific techniques. Analytical challenges to quantitation include the presence of interfering lipids, molecular instability, and extremely low physiological levels. High performance liquid chromatography (HPLC) with fluorescence and electrochemical detection (FLD-ECD) methods used for vitamin K1 involve labor-intensive, liquid-liquid extraction procedures and extremely long run times. Our goal was to develop and validate a high throughput and accurate method for measuring vitamin K1 in plasma and serum using liquid chromatography with tandem mass spectrometry (LC-MS/MS). Methods Plasma or serum specimens were mixed with stable isotope internal standard in an extraction solution containing acetonitrile to release vitamin K1 and precipitate proteins. After centrifugation, the supernatant was then extracted using a HybridSPE® (Supelco) plate to remove the remaining precipitated proteins and phospholipids that may contribute to ion-suppression. Certified standards were used to derive a 6-point calibration curve. Extracts (plasma or serum) and calibrators were analyzed on a SCIEX 6500 + LC-MS/MS with a Thermo TX-4 HTLC multiplexed 4-channel system for increased throughput. Chromatographic separation was achieved on a Biphenyl analytical column. Precision, accuracy, linearity, recovery, and stability were evaluated. Random specimen discards, previously submitted for vitamin K1 testing, were used for a split-sample comparison study vs the current HPLC-FLD-ECD method. Results Compared with the HPLC-FLD-ECD method, the LC-MS/MS method employed a simplified automated hybrid solid phase extraction procedure, used less specimen volume (100 µL vs 700 µL), increased total throughput (4 h vs 32 h per 96-well plate), was more sensitive (limit of quantitation [LOQ] 6.1 pg/mL vs 25.0 pg/mL), and provided greater molecular specificity for vitamin K1 through multiple reaction monitoring (MRM). The intra-assay and total inter-assay precision coefficients of variation (CVW, CVT) in plasma samples (n = 100) were: (6.8%, 5.8%) at 100 pg/mL, (3.5%, 3.0%) at 800 pg/mL, and (3.0%, 3.9%) at 2000 pg/mL. In spiked plasma and serum samples (n = 10), the linearity range was 40–20 000 pg/mL and percent recovery were 98%–107%. Least-squares regression analysis comparing the results for split plasma extracts by LC-MS/MS vs HPLC-FLD-ECD (n = 100) yielded correlation coefficients for vitamin K1: 0.9917, y = 0.927x + 0.827, range 40–7000 pg/mL. Vitamin K1 was stable in plasma and in serum stored in amber plastic screw-capped vials for 8 days at 2–8 °C and 30 days at −20 °C. Conclusion We have developed and validated an LC-MS/MS method that provides rapid, sensitive, precise, and accurate evaluation of vitamin K1 in plasma and serum. Advantages over the existing HPLC-FLD-ECD method include increased throughput and improved sensitivity and specificity. Vitamin K1 results obtained using this method can be used by clinicians to guide clinical intervention to improve or normalize metabolism or prevent the potential adverse consequences associated with inadequate dietary intake.