Development of vitamin K measurement method and determination of reference range

Vitamin K plays a crucial role in essential metabolic processes such as post-transcriptional modification of glutamic acid residues, bone mineralization, calcium homeostasis, and cell cycle regulation in the human body. Therefore, determining the level of vitamin K is of great importance in diagnosing complications related to its deficiency. In this doctoral thesis, the aim is to develop an LC-MS/MS measurement method with analytical performance characteristics compliant with standards for vitamin K and its derivatives, using an innovative extraction method that provides high recovery without solid-phase extraction. Additionally, the goal is to determine the reference range of vitamin K specific to the Turkish population. This will contribute to the determination of serum vitamin K levels in patients and the diagnosis of complications related to vitamin K deficiency. During the preparation of standards, Vitamin K-d7 (5,6,7,8-d4, 2-methyl-d3) was used as the internal standard. All standards were dissolved in methanol. The internal standard solution was prepared at a concentration of 500 ng/mL. For method calibration, standards were prepared with Phylloquinone and Vitamin K2 (MK-4) chemicals at concentrations ranging from 500 ng/mL to 0.06 ng/mL. Each calibration standard was supplemented with the internal standard to a final concentration of 10 ng/mL. During the preparation of serum samples, the internal standard prepared with Vitamin K-d7 (5,6,7,8-d4, 2-methyl-d3) was added first. Ethanol, methanol, acetonitrile, chloroform, n-hexane, lipase, and ammonium sulfate solutions were used to precipitate proteins in the serum. Incubation was performed in a hot water bath (37 °C) and a deep freezer (-20 °C) for 15-30 minutes, followed by vortexing and centrifugation. The upper phase was collected and evaporated using a concentrator device. The residue was dissolved in 2-propanol, acetonitrile, and methanol chemicals to prepare analytes for serum samples. The prepared analytes were analyzed using liquid chromatography-high resolution mass spectrometry. Atmospheric pressure chemical ionization (APCI) was used as the ionization source in the mass spectrometer. Chromatographic separation was performed using a C18 - 100 x 2 mm - 3 µ column at 40 °C. The mobile phase consisted of 0.1% formic acid in water and a mixture of methanol/acetoneitrile (3:1, v/v). The developed final method underwent validation studies including tests for limit of detection, linearity, accuracy, precision, and robustness. In the limit of detection tests, analytes of Vitamin K, MK-4, and MK-7 with a concentration of 0.5 ng/mL were analyzed in 10 replicates. The obtained results were multiplied by three times the standard error to determine the detection limit. The standard error was multiplied by ten to establish the quantification limit. For the linearity tests, calibration standards were prepared at 10 different concentrations. The prepared standards were analyzed in six replicates, and a calibration curve was plotted. The results were analyzed through regression analysis. In the accuracy, precision, and robustness tests, Vitamin K analytes at concentrations of 5, 10, and 20 ng/mL were prepared. The analytes were prepared by two different analysts on two separate days and analyzed in 10 replicates. The analysis results were used to calculate intra-day and inter-day average and relative standard deviation. In the recovery tests, a serum pool sample was prepared and divided into two portions. One of the portions was spiked with a 1 ng/mL Vitamin K solution. The analytes were analyzed in 10 replicates. The obtained results were used to calculate percent recovery and bias. Serum samples from a total of 64 healthy female and 67 healthy male control group volunteers who met the inclusion and exclusion criteria were collected to determine the reference range for Vitamin K in the Turkish population. The collected samples were prepared and analyzed according to the final sample preparation procedure of the developed method. Statistical analyses performed during the validation studies were carried out using the R programming language in the RStudio software, utilizing the dplyr, tidyr, and ggplot2 packages. Spectrum analyses were conducted using the open-source software MZmine 3. Reference range calculations were performed using the Python programming language on the Google Colab platform, employing the pandas, numpy, matplotlib, and seaborn libraries. In the selected method for The calibration standard for Vitamin K1 was performed with 6 replicates, resulting in a correlation coefficient of 0.9957. During the intra-day analysis, it was determined that the analytes stored at room temperature and 2-8°C did not pose any issues. However, due to the light sensitivity of Vitamin K and its derivatives, storage conditions should be kept dark. Vitamin K, MK-4, and MK-7 analytes were analyzed with 10 replicates, and the standard error of the results was calculated as 0.009. Based on this, the method's limit of detection (LOD) and limit of quantification (LOQ) were calculated as 0.02 ng/mL and 0.09 ng/mL, respectively. The average relative standard deviation was determined as 2.15%. In the recovery tests, the levels of Vitamin K, MK-4, and MK-7 in the prepared serum pool sample were found to be 1.72 ng/mL, 0.84 ng/mL, and 0.53 ng/mL, respectively. An additional 1 ng/mL of Vitamin K was added to the sample, and after analyzing with 10 replicates, the average percent recovery values for Vitamin K, MK-4, and MK-7 were calculated as 94%, 88%, and 85%, respectively. For the determination of the reference range, a reference population consisting of 64 healthy female and 67 healthy male control group volunteers who met the inclusion and exclusion criteria was included. As a result, the reference range was calculated as 0.1-1.2 ng/mL. The method can provide results within a maximum of 30 minutes, including the pre-processing steps. Its distinguishing features from other methods in the literature include the absence of solid-phase extraction filters during the pre-processing step and the ability to deliver rapid results. After conducting method validation studies in research laboratories, it can be used for its intended purpose. Once method comparison and method validation studies are completed, it can be implemented for routine use. Python programming language was used to write the code for determining the reference range of Vitamin K, MK-4, and MK-7. To establish laboratory-specific reference ranges, which are recommended in clinical laboratories, this code can be integrated into laboratory information systems for all parameters. Enzyme-assisted extraction can also be utilized in the development of methods for quantifying other fat-soluble vitamins. Further research studies can be conducted in this field.