MSACL 2022 Abstract

Background
Serum 25-hydroxy vitamin D (25OHD) is widely used as a biochemical marker of vitamin D status. Most 25OHD circulating in blood is tightly bound to vitamin D binding protein, a smaller fraction is weakly bound to albumin, and a very small fraction (<0.1%) is present in a free form. It has been demonstrated that most cells in the human body respond to free rather than bound 25OHD and as a result, concentrations of free 25OHD (F25OHD) are likely more relevant than total 25OHD (T25OHD) for the assessment of vitamin D status. We developed a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for the measurement of F25OHD2 and F25OHD3 and used the method in a study of the effect of percent F25OHD on 25OHD clearance in individuals with and without kidney function impairment.

Methods
Sample preparation in the method was performed as follows: F25OHD2/ F25OHD3 were separated from the protein bound fraction using size exclusion-based separation; the analytes and the corresponding internal standards were derivatized using Ampliflex diene reagent (Sciex), and the samples were analyzed by LC-MS/MS. The instrumental analysis utilized two-dimensional chromatographic separation with data acquisition performed on a Triple QuadTM 6500 (Sciex). The lower limit of quantification of the assay for F25OHD2 and F25OHD3 was 0.005 ng/mL, and the total imprecision was <18%. Reasonably good correlation (r=0.873, n=62) was observed with a commercial ELISA (DIAsource), while concentrations measured by the ELISA were significantly lower than by the LC-MS/MS method. In the study on the effect of percent F25OHD on 25OHD clearance, we determined the serum T25OHD concentrations using LC-MS/MS; and F25OHD concentrations using a) LC-MS/MS, b) commercial ELISA (DIAsource), and c) calculation-based method, in samples of healthy adults (n=42), adults with chronic kidney disease (n=24) and kidney failure (n=19). We assessed agreement among the 3 methods and the association of F25OHD concentrations with the kidney function.

Results
We observed reasonably good correlation between the LC-MS/MS method and the commercial ELISA (r=0.61, n=42) in samples of healthy adults, while the concentrations by the ELISA were on average 10 times lower as compared to the LC-MS/MS method. Significant negative biases were observed in the ELISA measured F25OHD concentrations among participants with kidney failure, as compared to the LC-MS/MS, suggesting potential sample matrix interference with the ELISA measurements in these samples. In samples of healthy adults, we observed reasonably good correlation between the measured by LC-MS/MS and the calculated concentrations of F25OHD (r=0.83, n=42). The percent F25OHD determined by the calculation-based method was weakly associated with 25OHD clearance, while this association was absent when percent F25OHD was determined using LC-MS/MS-measured F25OHD and T25OHD concentrations.

Conclusions
In conclusion, sensitivity and specificity of the LC-MS/MS method for F25OHD were adequate for measurement of endogenous F25OHD concentrations in samples from healthy individuals and patients with kidney disease. Our data suggest that percent F25OHD is not strongly associated with 25OHD clearance and that the percent F25OHD is not altered in patients with kidney failure as compared to individuals with normal kidney function. We determined that, when F25OHD concentrations were measured by the commercial ELISA, the apparent decrease in percent F25OHD in association with kidney failure was an artifact caused by interference in the ELISA.