VU Research Portal

VU Research Portal

The When, What & How of Measuring Vitamin D Metabolism

Dirks, N.F.

2020

document version

Publisher's PDF, also known as Version of record

Link to publication in VU Research Portal

citation for published version (APA)

Dirks, N. F. (2020). The When, What & How of Measuring Vitamin D Metabolism.

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal ?

Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

E-mail address:

vuresearchportal.ub@vu.nl

Part IV

Chapter 8

130

Chapter 8

Over the last couple of years, the LC-MS/MS has made its grand entry into our clinical laboratories [1]. Especially in the case of steroid hormones, it is considered a substantial improvement, because the traditionally used immunoassays suffered from lack of specificity. As such, an increasing number of hospital laboratories worldwide have now transferred to LC-MS/MS methods for determination of vitamin D. Measuring the metabolites and regulators of vitamin D with LC-MS/ MS opens up many more opportunities than possible with the immunoassay. This dissertation focusses on the most important markers in vitamin D assessment: 25-hydroxyvitamin D (25(OH)D), 1,25-dihydroxyvitamin D (1,25(OH)₂D), 24,25-dihydroxyvitamin D (24,25(OH)₂D) and fibroblast growth factor 23 (FGF23). It aims to improve methodology for biomarkers of vitamin D metabolism and shows their utility in clinical diagnostics and research.

Part II of this dissertation focusses on the methodology. It studies the quality of

the existing methods and provides new methods that may be applied in diagnostics and research.

25(OH)D

Many laboratories nowadays run 25(OH)D LC-MS/MS methods as an improvement over their previously used immunoassays. Unfortunately, the variation in results generated by the various 25(OH)D LC-MS/MS methods is still large. Therefore, the focus now lies on method standardization to facilitate comparison of results generated by different LC-MS/MS methods. In Chapter 2, we investigated the

current standardization status of the routine 25(OH)D LC-MS/MS methods employed in the Netherlands by comparing them with each other and with the reference measurement procedure. Although the LC-MS/MS methods differed substantially in their calibration procedures, the generated results in healthy subjects (R ≥ 0993) and in various patient samples (R ≥ 0989) highly correlated. We did observe some small relative biases (<14%) in both comparisons for two of the methods that may be resolved by recalibration of the methods concerned. We concluded that 25(OH)D measurement in the Netherlands is, despite substantial

131

Summary difference in calibration procedures, well standardized. Contrarily, international DEQAS results for 25(OH)D LC-MS/MS assays show a small average bias from the target value, but a relatively large CV of all LC-MS/MS method results compared to the results from the most-used immunoassays (unpublished results). This means we have to continue standardization efforts worldwide and remain critical in our evaluation of the quality of our 25(OH)D LC-MS/MS methods.

1,25(OH)

D

Only a few LC-MS/MS methods have been developed for 1,25(OH)₂D measurement. As no reference measurement procedure has been published to date, standardization is a bridge too far at this moment. For now, we can only aim for harmonization.

Chapter 3 describes the characteristics of a newly established LC-MS/MS method

for measurement of 1,25(OH)₂D. After extensive validation, it was compared to a commercial immunoassay, another published LC-MS/MS assay from the Catholic University of Leuven and the results from the Vitamin D External Quality Assessment Scheme (2012/2013). The method operated by the Catholic University of Leuven provided 21% higher results. It was later found to be partly caused by the presence of 1β,25(OH)₂D₃, an interferencethat does not affect our method [2]. After optimization of the Leuven method that removed this interference, the bias with our method was reduced to 14%. Where the remnant difference originates from remains unclear but proves we have steps to take in improvement and harmonization of our 1,25(OH)₂D methods.

We have provided our method with accessory reference values. 1,25(OH)₂D₃ concentrations were determined in 96 healthy subjects, men and women, between 20 and 70 years of age. For 1,25-dihydroxyvitamin D₃, this resulted in an interval from 59 to 159 pmol/L. Without a reference measurement procedure and subsequent standardization of 1,25(OH)₂D methods, results need to be considered in light of the method used.

24,25(OH)

D

132

Chapter 8

yet well-established. Nevertheless, a candidate reference measurement procedure has been published and used to assign values to Standard Reference Material (SRM) 972a [3, 4]. A recent publication describing the interlaboratory comparison of six LC-MS/MS methods for determination of 24,25(OH)₂Ddemonstrates undesirable biases, and thus the need for method standardization [5]. Chapter 4 describes

our LC-MS/MS method for simultaneous measurement of 24,25(OH)₂D and 25(OH)D. It was validated according to FDA and EMA guidelines and calibrated using SRM 972a. Reference values were established by determining 24,25(OH)₂D concentrations and the 25(OH)D/24,25(OH)₂D ratio in 92 healthy subjects, men and women, between 20 and 70 years of age. This resulted in reference ranges of 0.4―8.9 nmol/L and 10-33 for 24,25(OH)₂D and the 25(OH)D/24,25(OH)₂, respectively. The method and reference ranges can be used to assess the efficiency of catabolic clearance of 25(OH)D by the responsible CYP24A1 enzyme. As such, it can identify patients with inactivating mutations in CYP24A1, as seen in idiopathic infantile hypercalcemia. As our method is calibrated using SRM 972a, the reference values can be universally applied.

Hopefully, DEQAS will soon accompany their samples with target values for 24,25(OH)₂D in order to assess standardization of all current 24,25(OH)₂D LC-MS/ MS methods across time.

fgf23

FGFG23 is not a steroid hormone, but a protein hormone, which means LC-MS/MS method development is far more difficult. Hence, to this day no LC-MS/MS methods have been described. And so, in Chapter 5 we have studied the contemporary

immunoassays for measurement of FGF23. More specifically, the presumed pre-analytical instability of intact FGF23 when using these assays [6]. By combining new experiments with findings from other recent publications, we have shown that the use of protease inhibitors—as has been advocated before—is not necessary when using any of the currently available commercial immunoassays measuring intact FGF23. We do, however, still recommend rapid centrifuging of samples to prevent stability issues, which may emerge after excessive delay of processing.

133

Summary Notwithstanding, using the LC-MS/MS for quantitation of FGF23 would in theory enable us to study the protein in more depth than ever possible with an immunoassay. Post-translational modifications, different protein fragments and mutations lead to a different protein mass and can thus be identified with a mass spectrometer. The difficulty lies in the complexity of the sample preparation (protein-enrichment, protein digestion steps), and sensitivity of our current mass-spectrometers. Whether the LC-MS/MS will also replace the immunoassay in the clinical laboratory in the case of protein hormones therefore remains to be seen. The studies in part II of this dissertation contribute to the quality and understanding

of our methodology for vitamin D assessment. As the different vitamin D metabolites and FGF23 are all in different phases of overall method improvement (Figure 1), they all required a different approach, which resulted in the diverse set of publications described in part II. The continuing efforts to improve our LC-MS/MS methods will also facilitate measurement in more complex sample matrices, such as dried blood spots or tissue. Over the next couple of years, many more LC-MS/ MS methods will be reported and the overall quality should ameliorate. At the same time, simplification and automation of our LC-MS/MS methods will enable more laboratories access to it, as it would lighten the demand for highly-skilled personnel. Hopefully, this process of simplifying and automation of the LC-MS/MS technology will not lead to undoing of the accomplished performance of our current methods. That would mean taking a step back with regard to method accuracy and specificity.

Part III of this dissertation describes applications of the 24,25(OH)₂D and 25(OH) D LC-MS/MS method in clinical diagnostics and research.

LC-MS/MS methods for measurement of the vitamin D metabolites are becoming more of a regularity in research studies. Now is the time to deploy the improved methodology of the last couple of years in favour of better understanding of the role of vitamin D in health and disease. New generations of improved, more sensitive LC-MS/MS instruments will enable us to further meticulously study, to date unexplained, vitamin D metabolism-related conditions and the beneficial effects of

134

Chapter 8

vitamin D sufficiency. The endless number of studies describing associations between decreased 25(OH)D concentrations in blood and the risks of disease—which have still mostly relied on immunoassays—have resulted in continues increment of the 25(OH)D concentration we define as deficient. By using values of >50 nmol/L or even >75 nmol/L to define vitamin D sufficiency, we deem a considerable amount of the apparently healthy population insufficient or ill. We might therefore consider the possibility that our understanding of the value of vitamin D sufficiency and the corresponding recommended 25(OH)D concentrations, have been clouded by using dubious and unstandardized immunoassays in the process. As the costs for implementing LC-MS/MS have decreased and throughput is improved, its use in larger studies is now becoming feasible, as chapters 6 and 7 show.

Chapter 6 describes a study in patients suffering from acute lymphoblastic leukaemia.

It shows an association between the decrease in 25(OH)D₃ concentrations during high-dose methotrexate treatment and the development of severe oral mucositis.

Phase 1: Development Phase 2: Comparison Phase 3: Harmonization Phase 4: Standardisation 25(OH)D First LC-MS/MS method Multiple LC-MS/MS methods Periodic quality assessment Reference measure-ment procedure Phase 1 Phase 2 Phase 3 Phase 4 1,25(OH)₂D First LC-MS/MS method Multiple LC-MS/MS methods Periodic quality assessment Reference measure-ment procedure Phase 1 Phase 2 Phase 3 Phase 4 24,25(OH)₂D First LC-MS/MS method Multiple LC-MS/MS methods Periodic quality assessment Reference measure-ment procedure Phase 1 Phase 2 Phase 3 Phase 4 FGF23 First LC-MS/MS method Multiple LC-MS/MS methods Periodic quality assessment Reference measure-ment procedure

Figure 1. Depiction of the different phases of overall method improvement for 25(OH)D,

135

Summary 25(OH)D₃ concentrations before the start of high-dose methotrexate treatment were not associated with developing methotrexate-induced oral mucositis. Whether the decrease in 25(OH)D₃ is an effect or a cause of the development of oral mucositis was not studied and remains unknown.

Chapter 7 shows the 25(OH)D/24,25(OH)₂D ratio has no added value in predicting the individual response to vitamin D supplementation compared to the predictive value of 25(OH)D alone.

In conclusion, this dissertation provides descriptions of three LC-MS/MS methods for measurement of the main vitamin D metabolites: 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D and 24,25-dihydroxyvitamin D. The LC-MS/MS methods described here, in combination with the established reference intervals, are suitable for use in diagnosis of several vitamin D metabolism related disorders and have proven to be useful in research. In addition, we showed that use of protease inhibitors, often logistically complex and expensive, is not necessary anymore when measuring intact FGF23, which makes studies regarding FGF23 much simpler. As a whole, this dissertation contributes to an improvement in the measurement and interpretation of the vitamin D metabolism and FGF23.

references

1. Annesley, T.M., R.G. Cooks, et al., Clinical Mass Spectrometry-Achieving Prominence in Laboratory Medicine. Clin Chem, 2016. 62(1): p. 1-3.

2. Pauwels, S., I. Jans, et al., 1beta,25-Dihydroxyvitamin D3: A new vitamin D metabolite in human serum. J Steroid Biochem Mol Biol, 2017. 173: p. 341-348.

3. Tai, S.S. and M.A. Nelson, Candidate Reference Measurement Procedure for the Determination of (24R),25-Dihydroxyvitamin D3 in Human Serum Using Isotope-Dilution Liquid Chromatography-Tandem Mass Spectrometry. Anal Chem, 2015. 87(15): p. 7964-70.

4. Tai, S.S., M.A. Nelson, et al., Development of Standard Reference Material (SRM) 2973 Vitamin D Metabolites in Frozen Human Serum (High Level). J AOAC

Int, 2017. 100(5): p. 1294-1303.

5. Wise, S.A., S.S. Tai, et al., Interlaboratory Comparison for the Determination of 24,25-Dihydroxyvitamin D(3) in Human Serum Using Liquid Chromatography with Tandem Mass Spectrometry. J AOAC Int, 2017. 100(5): p. 1308-1317.

6. Smith, E.R., M.L. Ford, et al., Instability of fibroblast growth factor-23 (FGF-23): implications for clinical studies. Clin Chim Acta, 2011. 412(11-12): p. 1008-11.