Quantification of 1α,25-dihydroxy vitamin D by immunoextraction and liquid chromatography-tandem mass spectrometry

Abstract

Background: 1α,25-dihydroxy vitamin D [1,25(OH)(2)D] is the active metabolite of vitamin D. Antibody-based detection methods lack specificity, but when combined with isotope dilution/ultra-performance liquid chromatography (UPLC)-tandem mass spectrometry, immunoextraction provides an attractive method for 1,25(OH)(2)D. We developed a method for simultaneous quantification of 1,25(OH)(2)D(2) and 1,25(OH)(2)D(3) with a 4.6-min instrument cycle time. Results are available 36 h after sample preparation begins.

Methods: Sample preparation consisted of protein precipitation, immunoextraction with solid-phase anti-1,25(OH)(2)D antibody, and derivatization with 4-phenyl-1,2,4-triazoline-3,5-dione. Analytes were resolved using reversed-phase UPLC and quantified using positive ion electrospray ionization-tandem mass spectrometry. We used hexadeuterated 1,25(OH)(2)D(3) and 1,25(OH)(2)D(2) as internal standards and performed method comparisons against the DiaSorin RIA and an LC-MS/MS method available at a reference laboratory.

Results: 1,25(OH)(2)D(3) intraassay and interassay imprecision was 5.6% and 8.0% (120 pmol/L) and 8.7% and 13% (48 pmol/L). Limits of detection and quantification were 1.5 pmol/L and 3.0 pmol/L, respectively. 1,25(OH)(2)D(2) intraassay and interassay imprecision was 8.7% and 11% (186 pmol/L) and 11% and 13% (58 pmol/L). Limits of detection and quantification were both 1.5 pmol/L. Comparison with RIA had a proportional bias of 0.75, constant bias of -4.1, and Pearson correlation (r(2)) of 0.31. Comparison with a reference LC-MS/MS assay had a proportional bias of 0.89, constant bias of 3.7, and r(2) of 0.88.

Conclusions: Protein precipitation with antibody-based extraction is effective for sample preparation before LC-MS/MS analysis of derivatized 1,25(OH)(2)D. This method appears to have improved specificity over a clinically used RIA with low imprecision and limits of detection.

Figure 1. Representative chromatograms of 1,25(OH)₂D₃, 1,25(OH)₂D₂ and a qualitative ion suppression analysis

Traces of relative ion intensity from sera with 120 pmol/L of 1,25(OH)₂D₃ and 1,25(OH)₂D₂ plus addition of 200 pmol/L of internal standards. The inset is a tracing of ion suppression. The thin black line indicates the relative intensity of an infusion of internal standards as described in Materials and Methods. Representative peaks for 1,25(OH)₂D₃ and 1,25(OH)₂D₂ are filled in black and labeled as 1 and 2, respectively.

Figure 2. Vitamin D metabolites immunoextracted from a human sample by the anti-1,25(OH)₂D antibody

Chromatograms from an analysis of immunoextracted patient sera modified to include transitions corresponding to 24,25(OH)₂D₃ (574.34 > 298.12) and 25(OH)D₃ (558.34 > 298.12). 24,25(OH)₂D₃ and 25(OH)D₃ are well separated from both 1,25(OH)₂D₃ and 1,25(OH)₂D₂. The internal standard is shown for 1,25(OH)₂D₂ due to an undetectable amount of endogenous analyte in this sample.

Figure 3. Method comparison of the new LC-MS/MS assay with a 1,25(OH)₂D radioimmunoassay and reference LC-MS/MS method

(Left) Total 1,25(OH)₂D (pmol/L) measured with the DiaSorin RIA compared to total 1,25(OH)₂D measured by LC-MS/MS (n=116). The black dotted line is the line of identity and the solid black line is the Deming regression line (−4.1 + 0.76×; r² (Pearson) = 0.31). (Right) Comparison with a reference LC-MS/MS assay had a porportional bias of 0.89, constant bias of 3.7 and Pearson correlation (r²) of 0.88. (Bottom) Bland-Altman plots of the average concentration against the relative difference for RIA versus LC-MS/MS and LC-MS/MS versus reference LC-MS/MS results. The mean difference with ± 2 SD of relative differences is shown (dashed lines).