Development and optimization of simplified LC–MS/MS quantification of 25-hydroxyvitamin D using protein precipitation combined with on-line solid phase extraction (SPE)

25-Hydroxyvitamin D, the most useful marker of the vitamin D status of an individual, has seen an exponential growth of its routine measurement in recent years. Several methods are currently offered but the most specific is LC–MS/MS. However, the routine use of this technique in the clinical laboratory makes it essential to improve key steps of this method for high throughput delivery. Importantly, the preanalytical steps of this assay and the efficacy of the separation system need to be optimized prior to MS detection. In this report we replaced the standard and time consuming liquid–liquid extraction method of vitamin D metabolites with hexane (LLE) combined with centrifugation (LLE/centrifugation) by a simpler protein precipitation with extraction (PPE) in acetonitrile combined with a fast separation process using a 96-well plate filtration system (PPE/filtration). This rapid extraction was then followed by an on-line solid phase extraction (SPE) using a selective chromatographic separation. We also optimized the operational and consumable costs, by using an inexpensive guard column as a trapping column to significantly enhance the lifespan of the analytical column two to three times as compared to conventional chromatography. The LC–MS/MS technique permits the measurement of both 25-hydroxyvitamin D2 (25-OH D2) and the 25-hydroxyvitamin D3 (25-OH D3) metabolites in electrospray ionization (ESI) mode. The chromatographic system consisted of a 2.1 mm × 50 mm C18 3.5 μM column with a 2.1 mm × 20 mm C18 3.5 μM guard column connected with two 6 ports switching valves. Quantifications were done using the isotopic dilution technique with hexadeutered 25-OH D3 and 25-OH D2.The ion suppression problem with phospholipids was also evaluated and optimized to minimize this effect through the chromatography process and the on-line SPE trapping. Calibration curves were prepared by diluting a commercial high calibrator Chromsystems (München, Germany) with either pure triple stripped blank serum or diluted in 6% phosphate buffer saline at pH 7.2. Linearity was tested up to 160 nmol/L for 25-OH D3 and 75 nmol/L for 25-OH D2. Low limit of quantification (LLOQ) were established at 3 nmol/L for 25-OH D2 and 4 nmol/L for 25-OH D3. Inter-assay and intra-assay precision (CV%) was determined using 3 levels of commercial controls (Utak, CA, USA) for 25-OH D2 and 25-OH D3. Results obtained for intra-assay and inter-assay precision (CV%) were 1.1–3.4% and 5–8.9% respectively for the PPE/centrifugation technique and 2.0–3.1% and 4.6–6.6% for the PPE/filtration technique. Accuracy was estimated with the same commercial controls: % bias was −11.2 to 4.9% with PPE/centrifugation and −3.2 to 6.1% with PPE/filtration. 25-OH D2 and 25-OH D3 concentrations in human serum with LLE were compared to the new extraction methods using either PPE/centrifugation or PPE/filtration. Correlations comparing the two methods revealed a slope approximately 1.0 ± 0.3 with R ≥ 0.98 with a bias < 1 nmol/L. In summary, the new LC–MS/MS method described in this report using an on-line SPE technique with a simple off-line pre-treatment is faster, cost-effective, more reliable and more robust than current and widely used LLE/centrifugation methods coupled with LC–MS/MS.