Skip to main content
Article Publicly Available

Evaluation of two fully automated immunoassay based tests for the measurement of 1α,25-dihydroxyvitamin D in human serum and comparison with LC-MS/MS

  • EMAIL logo and
Published/Copyright: February 28, 2017
Download Article (PDF)
Clinical Chemistry and Laboratory Medicine (CCLM)
From the journal Clinical Chemistry and Laboratory Medicine (CCLM) Volume 55 Issue 9

Abstract

Background:

1α,25-Dihydroxyvitamin D [1,25(OH)2 vitD] is the bioactive form of vitamin D. Due to the very low concentrations of 1,25(OH)2 vitD in the blood and its lipophilic character, measurement of this parameter is analytically challenging. Requiring preceding manual extraction steps before analysis, previous assays have been laborious.

Methods:

In the presented study, we evaluated the performance of two immunoassays from DiaSorin and from Immunodiagnostic Systems (IDS) which combine fully automated extraction and measurement of 1,25(OH)2 vitD. Imprecision and linearity were verified according to Clinical and Laboratory Standards Institute EP15-A3 and EP6-A guidelines, respectively. Ninety-three patient serum samples sent to our institute for determination of 1,25(OH)2 vitD, as well as 20 Vitamin D External Quality Assessment Scheme (DEQAS) samples, were used to evaluate correlation and agreement of 1,25(OH)2 vitD measurements between the two immunoassays and with liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS).

Results:

Total imprecision was 5.2% or less for the DiaSorin test but reached 20.1% for the IDS iSYS test. 1,25(OH)2 vitD concentrations measured with the DiaSorin assay showed a strong correlation with 1,25(OH)2 vitD levels measured by LC-MS/MS and a good agreement with method specific means of DEQAS samples. By contrast, the IDS iSYS test overestimated 1,25(OH)2 vitD concentrations in human serum, particularly at higher concentrations.

Conclusions:

Due to its high sensitivity, low imprecision, broad measurement range, and good agreement with 1,25(OH)2 vitD concentrations measured by LC-MS/MS, the DiaSorin test is a valuable analytical option for the determination of 1,25(OH)2 vitD.

Keywords: 1α,25-dihydroxyvitamin D; immunoassay; method comparison; serum

Introduction

1α,25-Dihydroxyvitamin D [1,25(OH)2 vitD], the active form of vitamin D, is synthesized by mitochondrial 25-hydroxyvitamin D-1α-hydroxylase in the kidney and in extrarenal tissues. In addition to the kidney, particularly disease-activated tissue macrophages and the placenta may contribute to circulating concentrations of 1,25(OH)2 vitD [1]. 1,25(OH)2 vitD mediates actions generally ascribed to vitamin D including the absorption of calcium and phosphorus from the intestine, the retention of calcium from the kidney and bone mineralization, thereby preventing rickets and osteomalacia. Recently observed associations of low vitamin D levels with increased risks of cardiovascular diseases [2], [3], [4], type 2 diabetes [5], autoimmune diseases [6], infections of the upper respiratory tract [7], neurodegenerative disease [8], as well as breast or colorectal cancer [9], [10] point to extraskeletal actions of vitamin D. The expanded knowledge of the relevance of sufficient vitamin D levels is reflected by an experienced increased demand for the determination of vitamin D status in patients.

Although 1,25(OH)2 vitD is less suitable than 25-hydroxyvitamin D [25(OH) vitD] to assess vitamin D status, there are some indications to specifically determine 1,25(OH)2 vitD serum levels. This mainly includes differential diagnosis of hypercalcemia in patients with sarcoidosis and other granulomatous diseases. Determination of 1,25(OH)2 vitD can also be useful in patients with unexplained hyperparathyroidism who have adequate 25(OH) vitD levels. 1,25(OH)2 vitD is particularly suitable to differentiate between two hereditary defects namely vitamin D-dependent rickets (VDDR) type I, caused by vitamin D 1α-hydroxylase deficiency, and VDDR type II, caused by mutation of the vitamin D receptor gene leading to end-organ resistance to vitamin D. The growing knowledge of the prognostic significance of 1,25(OH)2 vitD after cardiac surgery and in sepsis and heart failure [11], [12], [13] will further increase the medical need to analyze 1,25(OH)2 vitD.

Due to its low concentration (pmol/L) and lipophilic nature, the quantification of 1,25(OH)2 vitD in serum is analytically challenging and requires extraction and separation steps prior to measurement. So far, mainly radioimmunoassays have been used for the measurement of 1,25(OH)2 vitD. They measure this parameter with sufficient sensitivity but are laborious, time consuming and prone to laboratory error associated with manual handling of the probes. The first automated assay including manual extraction had been developed by Immunodiagnostic Systems (IDS) Ltd., using the IDS iSYS immunoanalyzer. Recently, two assays became commercially available, which include not only the automated measurement but also an automated extraction of 1,25(OH)2 vitD prior to quantification. In this study, we evaluated and compared the performance of these two fully automated assays with each other and with liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS).

The measurement of 1,25(OH)2 vitD is poorly standardized. Whereas for the standardization of 25(OH) vitD measurements, a reference measurement procedure as well as National Institute of Standards and Technology standard reference material is available [14], [15], for 1,25(OH)2 vitD, a lot of work remains to be done to render the results from different laboratories and methods comparable. The Vitamin D External Quality Assessment Scheme (DEQAS) supports this standardization process by providing external quality control material not only for 25(OH) vitD but also for 1,25(OH)2 vitD. We therefore compared the values measured using the fully automated assays with the results obtained by DEQAS.

Materials and methods

Samples

Remaining material of 142 patient serum samples submitted to the Institute of Clinical Chemistry of the University Hospital Zurich for routine measurement of 1,25(OH)2 vitD was aliquoted and stored at −80 °C until analysis. All samples with extensive hemolysis, bilirubinemia or lipemia were excluded from the study. For analysis with the different tests, serum samples were thawed and one aliquot each was analyzed according to the manufacturer’s instructions. Of 93 samples, sufficient serum material was available for measurement of 1,25(OH)2 vitD concentrations using a LC-MS/MS based method.

Only samples of patients aged 18 years or older were included in the study. The study has been approved by the Zurich Cantonal Ethical Committee.

Immunoassays

Serum samples were measured with the LIAISON XL 1,25 Dihydroxyvitamin D assay (DiaSorin, Stillwater, MN, USA) on the LIAISON XL platform and with the IDS-iSYS 1,25 VitDXp assay (IDS, Boldon, UK) on the IDS-iSYS Multi-Discipline Automated System. Both tests are chemiluminescence-immunoassays and both tests capture the analyte from the probe prior to measurement of analyte concentration. Whereas the DiaSorin test uses a recombinant fusion protein to capture 1,25(OH)2 vitD from the probe [16], the IDS test uses anti-1,25(OH)2 vitD antibody coated magnetic particles for immunopurification. After several washing steps, the purified analyte is measured by an immunoassay. The DiaSorin test determines the 1,25(OH)2 vitD concentration directly after addition of the conjugate and a starter reagent to induce the light reaction [16]. The resulting signal is directly proportional to the analyte concentration. The IDS test, in contrast, uses a competitive immunoassay for detection of 1,25(OH)2 vitD concentration, where a certain concentration of an 1,25(OH)2 vitD acridinium conjugate competes for the binding sites of a biotinylated sheep anti-1,25(OH)2 vitD antibody. Thus, the resulting signal is inversely proportional to the amount of the analyte in the probe.

The performance characteristics of the immunoassays are outlined in Table 1.

Table 1:

Assay specifications of the two commercial assays for determination of 1,25(OH)2 vitD used in this study.

AssayLIAISON XL 1,25 Dihydroxyvitamin D assayIDS-iSYS 1,25 VitDXp assay
AnalyzerLIAISON XLIDS iSYS
PrincipleCLIA with prior extractionCLIA with prior extraction
Extraction of 1,25(OH)2 vitDRecombinant fusion proteinAnti-1,25(OH)2 vitD antibody
Antibody used for 1,25(OH)2 vitD determinationMouse, monoclonalSheep, monoclonal
LabelLuminolAcridinium
Measuring range5.0–200.0 ng/L7.5–210.0 ng/L
LoQ5.0 ng/LNot given
Needed amount of serum225 μL (75 μL+150 μL dead volume)400 μL (220 μL+180 μL dead volume)
Cross reactivity
 1,25(OH)2 vitD3100% (100 ng/L)98% (conc. not given)
 1,25(OH)2 vitD2104% (100 ng/L)57% (conc. not given)
 25(OH) vitD3<0.1% (100,000 ng/L)0.0014% (conc. not given)
 25(OH) vitD2<0.1% (50,000 ng/L)0.0005% (conc. not given)
 24,25(OH)2 vitD3<0.1% (50,000 ng/L)0.0005% (conc. not given)
 24,25(OH)2 vitD2<0.1% (50,000 ng/L)0.041% (conc. not given)
 3-epi25(OH) vitD3<0.1% (50,000 ng/L)0.0005% (conc. not given)

  1. Data derived from package inserts provided by the manufacturers. CLIA, chemiluminescence immunoassay; LoQ, limit of quantification.

Mass spectrometry

For measurement of 1,25(OH)2 vitD with LC-MS/MS, aliquots were shipped on dry ice to the laboratory of Prof. Hoofnagle at the University of Washington (Seattle, WA, USA). 1,25(OH)2 vitD2 and 1,25(OH)2 vitD3 serum concentrations were measured there as previously described [17].

Imprecision and accuracy

Assay imprecision has been estimated according to Clinical and Laboratory Standards Institute (CLSI) EP15-A3 guidelines [18]. Manufacturers’ quality control material as well as patient serum samples have been analyzed over 5 days with five measurements a day. From the obtained 25 measurements, within-run, between-run, and total imprecision have been estimated.

Linearity

Serial dilutions of a sample with high 1,25(OH)2 vitD concentration have been prepared using serum with very low or even non-measureable low levels as diluent in order to cover the measurement range of the tests. 1,25(OH)2 vitD concentrations were determined in duplicates and measured serum concentrations were compared to expected concentrations. Polynomial regression analysis has been used for the assessment of linearity according to the National Committee on Clinical Laboratory Standards (NCCLS) EP6-A evaluation protocol [19]. A linear or non-linear coefficient was assumed to be statistically significant if p was <0.05.

Method comparison

Frozen aliquots were thawed and equilibrated to room temperature. Samples were mixed gently and analyzed in batches on the IDS iSYS and the LIAISON XL immunoanalyzer. Quality control samples were analyzed together with each batch of samples to monitor the performance of the assays and the instruments. Correlation and agreement between the tested methods were assessed by means of Passing-Bablok regression analysis [20], Bland-Altman difference plots [21], and determination of Pearson’s coefficient of correlation. For correlation analysis, samples below the lower limit of the measurement range were assigned the value of the lower limit (7.5 and 5.0 ng/L for the IDS test and the DiaSorin test, respectively).

Statistical analysis

All statistical analyses were performed using Analyse-it, a statistical software for Microsoft Excel (Analyse-it Software Ltd.)

Results

Imprecision and accuracy

For the determination of imprecision, quality control material available from the respective manufacturer as well as human serum pools were analyzed according to the CLSI EP15-A3 guidelines. Imprecision results from a total of 25 measurements of each probe are shown in Table 2.

Table 2:

Imprecision and inaccuracy of the IDS iSYS and the DiaSorin LIAISON XL- 1,25(OH)2 vitD assays.

SampleHuman serumControl 1Control 2
Pool 1Pool 2Pool 3Pool 4
IDS iSYS
 Mean concentration, ng/L14.427.3108.143.288.6
 Target concentration, ng/L (range)35.6 (18.5–52.7)77.7 (49.7–105.7)
 Bias, ng/L7.610.9
 Bias, %21.314.1
 Imprecision, %
  Total (within laboratory)20.118.97.115.98.9
  Within run (repeatability)19.811.15.87.55.6
  Between run3.515.34.114.07.0
DiaSorin LIAISON XL
 Mean concentration, ng/L8.513.029.891.931.2121.9
 Target concentration, ng/L (range)30.4 (21.9–38.9)122.0 (92.8–151)
 Bias, ng/L0.8−0.1
 Bias, %2.5−0.1
 Imprecision, %
  Total (within laboratory)4.45.24.13.14.53.3
  Within run (repeatability)2.54.43.71.53.52.5
  Between run3.62.81.82.72.92.1

  1. For measurement of imprecision in human serum human serum pools were used. Controls 1+2: quality control material level 1 and 2 of the respective test. There were no statistical outliers as tested by Grubbs’ test.

The DiaSorin test determined 1,25(OH)2 vitD very precisely, with total imprecision ranging between 3.1% and 5.2% and hence within the range of 3.6%–6.6% reported by the manufacturer. In contrast, the total imprecision of the IDS test ranged between 7.1% and 20.1%. Intra- and inter-assay imprecisions were between 5.6% and 19.8% and between 3.5% and 15.3%, respectively, as compared to intra-assay and inter-assay imprecisions reported by the manufacturer, being 6.4%–12.1% and 4.6%–9.6%, respectively. However, for the imprecision determined by the manufacturer, only serum samples with concentrations of 25.3 ng/L and above have been investigated. Furthermore, we found that the DiaSorin test measured 1,25(OH)2 vitD concentrations with better accuracy than the IDS test, namely with a bias of 2.5% and −0.1% as compared to 21.3% and 14.1%.

Linearity

A series of dilutions was prepared by adding different amounts of serum with very low or non-measurable low 1,25(OH)2 vitD concentrations to a serum sample with high 1,25(OH)2 vitD concentrations. Within the measurement ranges either test showed good correlations between measured and expected 1,25(OH)2 vitD concentrations (Figure 1). Statistical analysis revealed that none of the non-linear coefficients was statistically significant. Thus, for the IDS test as well as the DiaSorin test a linear equation models the data best. Reflecting the high imprecision, the IDS test showed poorer agreement between duplicate measurements.

Figure 1: Assessment of method linearity for the IDS iSYS and the DiaSorin 1,25(OH)2 vitD assay.For evaluation of the IDS test a serum sample with a mean concentration of 188.2 ng/L 1,25(OH)2 vitD was diluted with a serum sample with a 1,25(OH)2 vitD concentration below the detection limit of the test. For the DiaSorin test a serum probe containing 188.5 ng/L 1,25(OH)2 vitD was diluted with serum with a 1,25(OH)2 vitD concentration of 5.8 ng/L. Statistical analysis was performed according to the NCCLS EP6-A evaluation protocol.
Figure 1:

Assessment of method linearity for the IDS iSYS and the DiaSorin 1,25(OH)2 vitD assay.

For evaluation of the IDS test a serum sample with a mean concentration of 188.2 ng/L 1,25(OH)2 vitD was diluted with a serum sample with a 1,25(OH)2 vitD concentration below the detection limit of the test. For the DiaSorin test a serum probe containing 188.5 ng/L 1,25(OH)2 vitD was diluted with serum with a 1,25(OH)2 vitD concentration of 5.8 ng/L. Statistical analysis was performed according to the NCCLS EP6-A evaluation protocol.

Method comparisons

Correlation and agreement between the immunoassays

One hundred forty-two serum samples, sent for routine analysis of 1,25(OH)2 vitD to our institute, were measured with both, the LIAISON XL 1,25 Dihydroxyvitamin D assay and the IDS-iSYS 1,25 VitDXp assay (for correlation see Supplementary Figure 1). In the 93 patient samples with sufficient material for measurement with either method, the immunoassays and LC-MS/MS, 1,25(OH)2 vitD concentrations ranged from <7.5 to 202.5 ng/L if measured with the IDS test, from <5.0 to 126.0 ng/L if measured with the DiaSorin test, and from 3.0 to 118.6 ng/L if measured by LC-MS/MS. 1,25(OH)2 vitD concentrations measured with the IDS test tended to be higher than those measured with the DiaSorin assay (Figure 2A). This was most pronounced in the higher concentration range (above about 40 ng/L) as indicated by the respective Bland-Altman-Plot (Figure 2B). Over the whole measurement range, the scattering of the values was high as reflected by the 95% limits of agreement of the bias (±1.96*standard deviation [SD]): Bland-Altman analysis revealed a mean bias of 4.9%, the lower and upper limits of agreement were −67.5% and 77.4%, respectively.

Figure 2: Method comparison of the two fully automated immunoassays and a LC-MS based method for measurement of 1,25(OH)2 vitD.Left panel: Passing-Bablok regression analysis of the IDS with the DiaSorin test (A) and the IDS (C) and the DiaSorin test (E), respectively, with LC-MS as a reference method. The bold red solid line represents the regression line, the fine grey line represents the line of equality. Right panel: Bland-Altman plots for comparison of the IDS with the DiaSorin assay (B) and the IDS (D) and the DiaSorin assay (F), respectively, with LC-MS. The bold blue solid line indicates the relative mean bias, the dashed blue lines represent the upper and lower limit of agreement, respectively.
Figure 2:

Method comparison of the two fully automated immunoassays and a LC-MS based method for measurement of 1,25(OH)2 vitD.

Left panel: Passing-Bablok regression analysis of the IDS with the DiaSorin test (A) and the IDS (C) and the DiaSorin test (E), respectively, with LC-MS as a reference method. The bold red solid line represents the regression line, the fine grey line represents the line of equality. Right panel: Bland-Altman plots for comparison of the IDS with the DiaSorin assay (B) and the IDS (D) and the DiaSorin assay (F), respectively, with LC-MS. The bold blue solid line indicates the relative mean bias, the dashed blue lines represent the upper and lower limit of agreement, respectively.

Correlation and agreement of the immunoassays with the LC-MS/MS method

To check for accuracy, results obtained by the two immunoassays on 93 patient samples were compared with the results obtained by LC-MS/MS, a method propagated as the reference method for 1,25(OH)2 vitD measurement. Compared to the IDS test, the DiaSorin measurement results showed stronger correlations with the LC-MS/MS results (r=0.852, Figure 2C vs. r=0.967, Figure 2E). Moreover, the IDS test overestimated 1,25(OH)2 vitD serum concentrations especially in the higher concentration range, whereas the DiaSorin test showed good agreement. In the Bland-Altman analysis mean bias was 7.0% (95% LoA, −69.8% and 83.9%) for the IDS test but only 2.3% (95% LoA, −29.2% and 33.7%) for the DiaSorin test (Figure 2D, F).

Unlike the immunoassays, LC-MS/MS can distinguish between 1,25(OH)2 vitD2 and 1,25(OH)2 vitD3. The majority of the samples had very low 1,25(OH)2 vitD2 levels, only 6 of the 93 samples with total 1,25(OH)2 vitD concentrations between 7.8 and 85.1 ng/L measured with LC-MS/MS contained 5% or more 1,25(OH)2 vitD2. One of these samples had a 1,25(OH)2 vitD concentration below the detection limit when measured with the IDS test and was therefore excluded from statistical analysis for this test. The 1,25(OH)2 vitD concentrations of the samples with 1,25(OH)2 vitD2 concentrations between 5.8% and 76.6% of total 1,25(OH)2 vitD showed a mean deviation of 86.7 (IDS) and 98.2% (DiaSorin) from 1,25(OH)2 vitD3 measured by LC-MS/MS, but of only 3.7 (IDS) and 1.2% (DiaSorin), respectively, when compared to the sum of 1,25(OH)2 vitD2 and 1,25(OH)2 vitD3. This suggests nearly complete cross-reactivity of either immunoassay with 1,25(OH)2 vitD2.

Performance in external quality control measurements

To date, no international standard is available for the measurement of 1,25(OH)2 vitD. To compensate for this lack, we measured 20 samples from DEQAS, the largest vitamin D quality assessment program worldwide, with the two novel immunoassays to compare our measurements with the results of other laboratories (Table 3). Seventy-five percent and 95% of results measured by the IDS test and the DiaSorin test, respectively, are within 1 SD of the method specific mean. None of our results showed deviations of 2 or more SDs from the method mean. Thus, measurements in our laboratory are comparable to the results obtained by other laboratories using the IDS iSYS or the DiaSorin test.

Table 3:

Immunoassay measurement results and method means and standard deviations of DEQAS samples.

DEQAS sample IDIDS iSYSDiaSorinLC-MS/MS
Measurement Value, ng/LMethod mean (SD)b, ng/LnaMeasurement Value, ng/LMethod mean (SD)b, ng/LnaMethod mean (SD)b, ng/Lna
35119.322.6 (6.2)3221.022.8 (4.6)1725.1 (7.0)13
35243.542.6 (7.3)3238.837.3 (7.7)1741.6 (13.1)13
35370.864.6 (10.4)3252.752.6 (9.7)1755.4 (22.0)13
35469.861.8 (9.7)3249.748.1 (8.0)1746.3 (13.1)13
35558.858.7 (7.7)3257.353.5 (9.0)1749.9 (15.5)13
36137.837.3 (8.9)3537.835.7 (4.8)4738.0 (8.2)13
36228.925.8 (6.3)3530.824.9 (3.3)4729.6 (7.5)13
36355.248.6 (9.2)3543.543.3 (4.8)4745.0 (8.0)13
36465.852.8 (9.3)3551.647.2 (5.0)4744.5 (8.1)13
36537.132.3 (6.5)3537.234.6 (3.8)4732.7 (6.0)13
37124.232.5 (7.3)3734.229.8 (5.0)7438.0 (5.5)14
37260.760.9 (10.1)3755.252.6 (8.1)7452.0 (9.8)14
37329.737.2 (8.9)3737.136.0 (5.2)7436.0 (7.2)14
37458.867.1 (11.5)3755.451.5 (7.4)7459.4 (18.8)14
37545.047.9 (9.2)3758.060.0 (7.9)7446.0 (19.1)14
37647.964.6 (13.5)3664.662.6 (6.7)7054.2 (6.3)14
37744.953.2 (10.3)3648.348.4 (5.5)7049.4 (7.2)14
37811.518.5 (5.7)3618.417.7 (2.8)7023.1 (9.8)14
37942.547.4 (9.6)3645.447.3 (5.3)7041.5 (7.3)14
38031.743.0 (9.2)3629.931.4 (4.2)7034.5 (9.9)14

  1. an, number of participants; bSD, standard deviation.

Both immunoassays showed a good agreement of the quality control sample measurements with LC-MS/MS. Mean deviations of the IDS and DiaSorin measurement results from the mean of LC-MS/MS measurements were 2.1% and 1.8%, respectively. However, individual measurements with the IDS test deviated more strongly from the LC-MS/MS method (range, −51% to 50%) than measurements with the DiaSorin test (−20% to 26%), reflecting the higher imprecision of the IDS test as seen in the method comparison and the imprecision study (Figure 3).

Figure 3: Relative deviation of 1,25(OH)2 vitD concentrations measured with the IDS iSYS and the DiaSorin 1,25(OH)2 vitD assay, respectively, from LC-MS/MS method mean in DEQAS samples.The blue rhombs represent IDS values, grey squares represent the concentrations measured with the DiaSorin test.
Figure 3:

Relative deviation of 1,25(OH)2 vitD concentrations measured with the IDS iSYS and the DiaSorin 1,25(OH)2 vitD assay, respectively, from LC-MS/MS method mean in DEQAS samples.

The blue rhombs represent IDS values, grey squares represent the concentrations measured with the DiaSorin test.

Discussion

In the presented study, we evaluated the analytical performance of two new commercially available fully automated assays for the measurement of 1,25(OH)2 vitD. Both immunoassays allow for determination of 1,25(OH)2 vitD within a shorter time frame compared to radioimmunoassays and other tests with manual sample pretreatment. Overall, the DiaSorin test performed better in terms of accuracy, sensitivity and imprecision compared to the IDS iSYS test.

The DiaSorin test running on the LIAISON XL analyzer measured 1,25(OH)2 vitD with good sensitivity and high reliability. Precision was very good even in the low concentration range and comparable to precision values previously described. Van Helden and Weiskirchen measured 1,25(OH)2 vitD in control material and serum pools with concentrations between 22.1 and 184.7 ng/L and found a repeatability and intermediate precision between 1.0%–5.0% and 3.8%–7.1%, respectively [16]. In a different study, intra- and inter-assay imprecision ranged from 1.1% to 4.7% and from 3.4% to 7.2%, respectively, in controls and serum pools with concentrations between 25.5 and 180.4 ng/L [22]. As we demonstrate here, the DiaSorin test measures 1,25(OH)2 vitD very precisely even at concentrations <10 ng/L. Furthermore, 1,25(OH)2 vitD concentrations measured with this test showed a very good correlation and agreement with 1,25(OH)2 vitD levels determined with the LC-MS/MS reference method.

The IDS assay with automated extraction running on the IDS iSYS immunoanalyzer measured 1,25(OH)2 vitD less precisely compared to the DiaSorin test. At concentrations of 14.4 ng/L total imprecision amounted to 20%. The limit of detection as indicated by the manufacturer is 7.5 ng/L for the IDS test. Even the limit of quantification of 5.0 ng/L for the DiaSorin test is lower than the IDS detection limit, reflecting the high imprecision of the IDS test.

Furthermore, compared to the DiaSorin test, 1,25(OH)2 vitD concentrations measured with the IDS test showed weaker correlations and poorer agreement with those measured by LC-MS/MS. Both the comparison with LC-MS/MS measurements and the measurement of DEQAS quality controls revealed that the IDS test underestimates 1,25(OH)2 vitD concentrations below 40 ng/L and overestimates higher concentrations. IDS test results deviated from the LC-MS/MS method mean of DEQAS by −50% to +51% compared to −20% to +26% deviations of the DiaSorin test.

The two methods apply different fully automated extraction procedures prior to the measurement of 1,25(OH)2 vitD. Whereas the DiaSorin test uses a recombinant fusion protein to capture 1,25(OH)2 vitD, the IDS test purifies the analyte from the serum using an anti-1,25(OH)2 vitD antibody. One might argue that the fusion protein binds to 1,25(OH)2 vitD with higher specificity than the antibody, leading to the lower imprecision and the better correlation with LC-MS/MS of the DiaSorin test. However, our currently used IDS test with manual extraction of 1,25(OH)2 vitD shows significantly stronger correlation with the LC-MS/MS measurements than the fully automated IDS test (r=0.908 vs. r=0.852, Figure 4), although this was still weaker than that of the DiaSorin test (r=0.967). Assuming that the same antibody is used for the manual and automated extraction prior to measurement of 1,25(OH)2 vitD, the automated extraction steps of the IDS test need to be optimized.

Figure 4: Method comparison of the IDS test using manual immunopurification of 1,25(OH)2 vitD with LC-MS analysis.Left panel (A): Passing-Bablok regression analysis. The bold red solid line represents the regression line, the fine grey line represents the line of equality. Right panel (B): Bland-Altman plot. The bold blue solid line indicates the relative mean bias, the dashed blue lines represent the upper and lower limit of agreement, respectively.
Figure 4:

Method comparison of the IDS test using manual immunopurification of 1,25(OH)2 vitD with LC-MS analysis.

Left panel (A): Passing-Bablok regression analysis. The bold red solid line represents the regression line, the fine grey line represents the line of equality. Right panel (B): Bland-Altman plot. The bold blue solid line indicates the relative mean bias, the dashed blue lines represent the upper and lower limit of agreement, respectively.

1,25-Dihydroxyvitamin D2 is derived from dietary ergocalciferol, which is subsequently 25-hydroxylated in the liver and 1α-hydroxylated in the kidney. Both, vitamins D2 and D3 are comparably biologically active [23]. In order to correctly represent the biologically relevant amounts, the tests used for the determination of 1,25(OH)2 vitD should measure either form. The LC-MS/MS method used in our study differentiates between D2 and D3. Of the six serum samples with 1,25(OH)2 vitD2 concentrations above 5% of total 1,25(OH)2 vitD, five samples contained amounts between 1.3 and 8.3 ng/L, most likely due to a high content of vegetable food and/or due to substitution. In one patient with a total 1,25(OH)2 vitD concentration of 29.7 ng/L, the 1,25(OH)2 vitD2 proportion amounted to 76.6%. Since we did not have any information about lifestyle or medication, the reason for the high 1,25(OH)2 vitD2 level in this particular patient remains unclear. When we compared the 1,25(OH)2 vitD concentrations measured with the immunoassays versus LC-MS/MS results in samples with significant amounts of 1,25(OH)2 vitD2, we found a mean deviation of 3.7% and 1.2% for the IDS test and the DiaSorin test, respectively. Thus, we assume that both tests almost completely cross-react with 1,25(OH)2 vitD2. This is in accordance with the manufacturer’s data for the DiaSorin test. For the IDS test, a cross-reactivity of 57% is indicated by the manufacturer, but the respective tested concentration is not indicated. In our study, we only had six samples with 1,25(OH)2 vitD2 concentrations above 5% of total 1,25(OH)2 vitD. Further studies involving more samples with significantly elevated 1,25(OH)2 vitD2 levels are necessary to precisely evaluate the cross-reactivity of the tests with 1,25(OH)2 vitD2. No or only very low cross-reactivity is indicated by either manufacturer for vitamin D metabolites with less biological activity.

Measurement of 1,25(OH)2 vitD using fully automated tests is of significant advantage, particularly as the number of samples submitted for the determination of this parameter has increased over the recent years. The implementation of such tests allows for higher throughput and less hands-on time compared to the manual tests used so far. Together with the high sensitivity, low imprecision, the broad measurement range and the good agreement with 1,25(OH)2 vitD concentrations measured with LC-MS/MS, the DiaSorin test improves 1,25(OH)2 vitD diagnostics. The fully automated test performed on the IDS iSYS has a comparable broad measurement range but needs further improvement with respect to sensitivity, precision and standardization. The respective limitations have most recently been addressed by the supplier by a further development of the test, for which the presented study will serve as an ideal starting point and framework for the evaluation of critical performance factors.

Corresponding author: Dr. med. Katharina Spanaus, Institute of Clinical Chemistry, University Hospital of Zurich, Raemistr. 100, 8091 Zurich, Switzerland, Phone: +41 44 255 3473, Fax: +41 44 255 4590

Acknowledgments

We are grateful to DiaSorin and IDS for the generous provision of the 1,25(OH)2 vitD tests free of charge for this study. We thank Beata Jaworek from the Institute of Clinical Chemistry for the excellent technical assistance.

  1. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.

  2. Research funding: None declared.

  3. Employment or leadership: None declared.

  4. Honorarium: None declared.

  5. Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.

References

1. Adams JS, Hewison M. Extrarenal expression of the 25-hydroxyvitamin D-1-hydroxylase. Arch Biochem Biophys 2012;523:95–102.10.1016/j.abb.2012.02.016Search in Google Scholar PubMed PubMed Central

2. Anderson JL, May HT, Horne BD, Bair TL, Hall NL, Carlquist JF, et al. Intermountain Heart Collaborative (IHC) Study Group. Relation of vitamin D deficiency to cardiovascular risk factors, disease status, and incident events in a general healthcare population. Am J Cardiol 2010;106:963–8.10.1016/j.amjcard.2010.05.027Search in Google Scholar PubMed

3. Mullie P, Autier P. Relation of vitamin d deficiency to cardiovascular disease. Am J Cardiol 2011;107:956.10.1016/j.amjcard.2010.10.079Search in Google Scholar PubMed

4. Sun Q, Pan A, Hu FB, Manson JE, Rexrode KM. 25-Hydroxyvitamin D levels and the risk of stroke: a prospective study and meta-analysis. Stroke 2012;43:1470–7.10.1161/STROKEAHA.111.636910Search in Google Scholar PubMed PubMed Central

5. Afzal S, Bojesen SE, Nordestgaard BG. Low 25-hydroxyvitamin D and risk of type 2 diabetes: a prospective cohort study and metaanalysis. Clin Chem 2013;59:381–91.10.1373/clinchem.2012.193003Search in Google Scholar PubMed

6. Antico A, Tampoia M, Tozzoli R, Bizzaro N. Can supplementation with vitamin D reduce the risk or modify the course of autoimmune diseases? A systematic review of the literature. Autoimmun Rev 2012;12:127–36.10.1016/j.autrev.2012.07.007Search in Google Scholar PubMed

7. Bergman P, Lindh AU, Björkhem-Bergman L, Lindh JD. Vitamin D and respiratory tract infections: a systematic review and meta-analysis of randomized controlled trials. PLoS One 2013;8:e65835.10.1371/journal.pone.0065835Search in Google Scholar PubMed PubMed Central

8. Knekt P, Kilkkinen A, Rissanen H, Marniemi J, Sääksjärvi K, Heliövaara M. Serum vitamin D and the risk of Parkinson disease. Arch Neurol 2010;67:808–11.10.1001/archneurol.2010.120Search in Google Scholar PubMed PubMed Central

9. Abbas S, Linseisen J, Slanger T, Kropp S, Mutschelknauss EJ, Flesch-Janys D, et al. Serum 25-hydroxyvitamin D and risk of postmenopausal breast cancer–results of a large casecontrol study. Carcinogenesis 2008;29:93–9.10.1093/carcin/bgm240Search in Google Scholar PubMed

10. Gorham ED, Garland CF, Garland FC, Grant WB, Mohr SB, Lipkin M, et al. Optimal vitamin D status for colorectal cancer prevention: a quantitative meta analysis. Am J Prev Med 2007;32:210–6.10.1016/j.amepre.2006.11.004Search in Google Scholar PubMed

11. Zittermann A, Kuhn J, Ernst JB, Becker T, Larisch J, Dreier J, et al. Circulating 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D concentrations and postoperative infections in cardiac surgical patients: the CALCITOP-Study. PLoS One 2016;11:e0158532.10.1371/journal.pone.0158532Search in Google Scholar

12. Nguyen HB, Eshete B, Lau KH, Sai A, Villarin M, Baylink D. Serum 1,25-dihydroxyvitamin D: an outcome prognosticator in human sepsis. PLoS One 2013;8:e64348.10.1371/journal.pone.0064348Search in Google Scholar

13. Zittermann A, Fuchs U, Kuhn J, Dreier J, Schulz U, Gummert JF, et al. Parameters of mineral metabolism predict midterm clinical outcome in end-stage heart failure patients. Scand Cardiovasc J 2011;45:342–8.10.3109/14017431.2011.611250Search in Google Scholar

14. Tai SS, Bedner M, Phinney KW. Development of a candidate reference measurement procedure for the determination of 25-hydroxyvitamin D3 and 25-hydroxyvitamin D2 in human serum using isotope-dilution liquid chromatography-tandem mass spectrometry. Anal Chem 2010;82:1942–8.10.1021/ac9026862Search in Google Scholar

15. Stepman HC, Vanderroost A, Van Uytfanghe K, Thienpont LM. Candidate reference measurement procedures for serum 25-hydroxyvitamin D3 and 25-hydroxyvitamin D2 by using isotope-dilution liquid chromatography-tandem mass spectrometry. Clin Chem 2011;57:441–8.10.1373/clinchem.2010.152553Search in Google Scholar

16. van Helden J, Weiskirchen R. Experience with the first fully automated chemiluminescence immunoassay for the quantification of 1α, 25-dihydroxy-vitamin D. Clin Chem Lab Med 2015;53:761–70.10.1515/cclm-2014-0698Search in Google Scholar

17. Strathmann FG, Laha TJ, Hoofnagle AN. Quantification of 1α,25-dihydroxy vitamin D by immunoextraction and liquid chromatography-tandem mass spectrometry. Clin Chem 2011;57:1279–85.10.1373/clinchem.2010.161174Search in Google Scholar

18. Clinical and Laboratory Standards Institute (CLSI). User verification of precision and estimation of bias; approved guideline-third edition. CLSI document EP15-A3. Wayne, PA: Clinical and Laboratory Standards Institute, 2014.Search in Google Scholar

19. National Committee on Clinical Laboratory Standards (NCCLS). Evaluation of the linearity of quantitative measurement procedures: a statistical approach; approved guideline. NCCLS document EP6-A, Wayne, PA: National Committee on Clinical Laboratory Standards, 2003.Search in Google Scholar

20. Passing H, Bablok W. A new biomerical procedure for testing the equality of measurements from two different analytical methods. J Clin Chem Clin Biochem 1983;21:709–20.Search in Google Scholar

21. Bland JM, Altmann DG. Statistical methods for assessing agreement between two methods of clinical measurements. Lancet 1986;8:332–7.10.1016/j.ijnurstu.2009.10.001Search in Google Scholar

22. Valcour A, Zierold C, Podgorski AL, Olson GT, Wall JV, DeLuca HF, et al. A novel, fully-automated, chemiluminescent assay for the detection of 1,25-dihydroxyvitamin D in biological samples. J Steroid Biochem Mol Biol 2016;164:120–26.10.1016/j.jsbmb.2015.08.005Search in Google Scholar

23. Reinhardt TA, Ramberg CF, Horst RL. Comparison of receptor binding, biological activity, and in vivo tracer kinetics for 1,25-dihydroxyvitamin D3, 1,25-dihydroxyvitamin D2, and its 24 epimer. Arch Biochem Biophys 1989;273: 64–71.10.1016/0003-9861(89)90162-8Search in Google Scholar

Supplemental Material:

The online version of this article offers supplementary material (DOI: https://doi.org/10.1515/cclm-2016-1074).

Received: 2016-11-25
Accepted: 2017-1-10
Published Online: 2017-2-28
Published in Print: 2017-8-28

©2017 Walter de Gruyter GmbH, Berlin/Boston

Articles in the same Issue

  1. Frontmatter
  2. Editorial
  3. Mass spectrometry or immunoassay: est modus in rebus
  4. Reviews
  5. The use of liquid chromatography-tandem mass spectrometry for therapeutic drug monitoring of antibiotics in cancer patients
  6. Tackling serum folate test in European countries within the health technology assessment paradigm: request appropriateness, assays and health outcomes
  7. Genetics and Molecular Diagnostics
  8. Genetic diagnosis of α1-antitrypsin deficiency using DNA from buccal swab and serum samples
  9. General Clinical Chemistry and Laboratory Medicine
  10. Serum triglyceride measurements: the commutability of reference materials and the accuracy of results
  11. Variant peptide detection utilizing mass spectrometry: laying the foundations for proteogenomic identification and validation
  12. Evaluation of two fully automated immunoassay based tests for the measurement of 1α,25-dihydroxyvitamin D in human serum and comparison with LC-MS/MS
  13. Parallel diurnal fluctuation of testosterone, androstenedione, dehydroepiandrosterone and 17OHprogesterone as assessed in serum and saliva: validation of a novel liquid chromatography-tandem mass spectrometry method for salivary steroid profiling
  14. Determination of oxycodone and its major metabolites noroxycodone and oxymorphone by ultra-high-performance liquid chromatography tandem mass spectrometry in plasma and urine: application to real cases
  15. Identification and quantitation of phosphatidylethanols in oral fluid by liquid chromatography-tandem mass spectrometry
  16. Relationship between plasma and salivary melatonin and cortisol investigated by LC-MS/MS
  17. Paramagnetic micro-particles as a tool for rapid quantification of apixaban, dabigatran, edoxaban and rivaroxaban in human plasma by UHPLC-MS/MS
  18. Measurements of serum non-ceruloplasmin copper by a direct fluorescent method specific to Cu(II)
  19. The serum concentrations of leptin and MCP-1 independently predict low back pain duration
  20. Immunoassay screening in urine for synthetic cannabinoids – an evaluation of the diagnostic efficiency
  21. Cancer Diagnostics
  22. Study of kallikrein-related peptidase 6 (KLK6) and its complex with α1-antitrypsin in biological fluids
  23. Cardiovascular Diseases
  24. A candidate liquid chromatography mass spectrometry reference method for the quantification of the cardiac marker 1-32 B-type natriuretic peptide
  25. The natriuretic peptide MR-proANP predicts all-cause mortality and adverse outcome in community patients: a 10-year follow-up study
  26. CASZ1 loss-of-function mutation contributes to familial dilated cardiomyopathy
  27. Diabetes
  28. Evaluating new HbA1c methods for adoption by the IFCC and NGSP reference networks using international quality targets
  29. Infectious Diseases
  30. Analytical and diagnostic performance of two automated fecal calprotectin immunoassays for detection of inflammatory bowel disease
  31. Letters to the Editor
  32. Is fasting necessary for lipid profile determinations? Some considerations from the perspective of the clinical laboratory
  33. Precision of nonfasting lipid profiles should focus on clinical relevance rather than necessarily obtaining the least variation
  34. Triglyceride concentrations should be measured after elimination of free glycerol to exclude interindividual variations due to adiposity and fasting status
  35. Estimation of the reference interval for serum folate measured with assays traceable to the WHO International Standard
  36. Implausible elevation of peripheral thyroid hormones during therapy with a protein supplement
  37. Interference in Na+ measurements on the Siemens RAPIDPoint® 500 after nortriptyline intoxication: a case report
  38. Usefulness of maternal red cell antibodies to predict hemolytic disease of the fetus and newborn and significant neonatal hyperbilirubinemia: a retrospective study
  39. Improvement of the Sandell-Kolthoff reaction method (ammonium persulfate digestion) for the determination of iodine in urine samples
  40. Clinical use of targeted high-throughput whole-genome sequencing for a dengue virus variant
https://doi.org/10.1515/cclm-2016-1074
Keywords for this article
1α,25-dihydroxyvitamin D; immunoassay; method comparison; serum

Articles in the same Issue

  1. Frontmatter
  2. Editorial
  3. Mass spectrometry or immunoassay: est modus in rebus
  4. Reviews
  5. The use of liquid chromatography-tandem mass spectrometry for therapeutic drug monitoring of antibiotics in cancer patients
  6. Tackling serum folate test in European countries within the health technology assessment paradigm: request appropriateness, assays and health outcomes
  7. Genetics and Molecular Diagnostics
  8. Genetic diagnosis of α1-antitrypsin deficiency using DNA from buccal swab and serum samples
  9. General Clinical Chemistry and Laboratory Medicine
  10. Serum triglyceride measurements: the commutability of reference materials and the accuracy of results
  11. Variant peptide detection utilizing mass spectrometry: laying the foundations for proteogenomic identification and validation
  12. Evaluation of two fully automated immunoassay based tests for the measurement of 1α,25-dihydroxyvitamin D in human serum and comparison with LC-MS/MS
  13. Parallel diurnal fluctuation of testosterone, androstenedione, dehydroepiandrosterone and 17OHprogesterone as assessed in serum and saliva: validation of a novel liquid chromatography-tandem mass spectrometry method for salivary steroid profiling
  14. Determination of oxycodone and its major metabolites noroxycodone and oxymorphone by ultra-high-performance liquid chromatography tandem mass spectrometry in plasma and urine: application to real cases
  15. Identification and quantitation of phosphatidylethanols in oral fluid by liquid chromatography-tandem mass spectrometry
  16. Relationship between plasma and salivary melatonin and cortisol investigated by LC-MS/MS
  17. Paramagnetic micro-particles as a tool for rapid quantification of apixaban, dabigatran, edoxaban and rivaroxaban in human plasma by UHPLC-MS/MS
  18. Measurements of serum non-ceruloplasmin copper by a direct fluorescent method specific to Cu(II)
  19. The serum concentrations of leptin and MCP-1 independently predict low back pain duration
  20. Immunoassay screening in urine for synthetic cannabinoids – an evaluation of the diagnostic efficiency
  21. Cancer Diagnostics
  22. Study of kallikrein-related peptidase 6 (KLK6) and its complex with α1-antitrypsin in biological fluids
  23. Cardiovascular Diseases
  24. A candidate liquid chromatography mass spectrometry reference method for the quantification of the cardiac marker 1-32 B-type natriuretic peptide
  25. The natriuretic peptide MR-proANP predicts all-cause mortality and adverse outcome in community patients: a 10-year follow-up study
  26. CASZ1 loss-of-function mutation contributes to familial dilated cardiomyopathy
  27. Diabetes
  28. Evaluating new HbA1c methods for adoption by the IFCC and NGSP reference networks using international quality targets
  29. Infectious Diseases
  30. Analytical and diagnostic performance of two automated fecal calprotectin immunoassays for detection of inflammatory bowel disease
  31. Letters to the Editor
  32. Is fasting necessary for lipid profile determinations? Some considerations from the perspective of the clinical laboratory
  33. Precision of nonfasting lipid profiles should focus on clinical relevance rather than necessarily obtaining the least variation
  34. Triglyceride concentrations should be measured after elimination of free glycerol to exclude interindividual variations due to adiposity and fasting status
  35. Estimation of the reference interval for serum folate measured with assays traceable to the WHO International Standard
  36. Implausible elevation of peripheral thyroid hormones during therapy with a protein supplement
  37. Interference in Na+ measurements on the Siemens RAPIDPoint® 500 after nortriptyline intoxication: a case report
  38. Usefulness of maternal red cell antibodies to predict hemolytic disease of the fetus and newborn and significant neonatal hyperbilirubinemia: a retrospective study
  39. Improvement of the Sandell-Kolthoff reaction method (ammonium persulfate digestion) for the determination of iodine in urine samples
  40. Clinical use of targeted high-throughput whole-genome sequencing for a dengue virus variant
Downloaded on 25.4.2026 from https://www.degruyterbrill.com/document/doi/10.1515/cclm-2016-1074/html?lang=en
Scroll to top button