Standardization of vitamin D assays: art or science?

In this edition of the journal, Carter and Jones ¹ address the difficult question of how to standardize assays for 25-hydroxyvitamin D. This is a very important problem as we have seen an explosion of interest in the measurement of 25-hydroxyvitamin D in recent years. This has mainly been driven by the clinical interest in the role of 25-hydroxyvitamin D in metabolic bone disease and the increasing recognition of the contribution of 25-hydroxyvitamin D deficiency to the clinical problems experienced in elderly patients with osteoporosis. More recently, guidelines in relation to chronic kidney disease have driven requests for these measurements.

Recommendations regarding the diagnosis of vitamin D deficiency have appeared and the circulating concentration defining vitamin D deficiency has been the subject of significant debate. In the past, frank osteomalacia or rickets with obvious clinical signs and symptoms defined vitamin D deficiency. Population studies have correlated the concentration of parathyroid hormone (PTH) with total 25-hydroxyvitamin D and either defined an elevated PTH as indicative of vitamin D deficiency or more recently established the total 25-hydroxyvitamin D concentration that, after statistical manipulation, results in an inflection point in PTH increase resulting in ‘secondary’ hyperparathyroidism. ^2–5 Other studies have used high bone turnover, ⁶ decreased bone mineral density, and the abolishment of seasonal variation in PTH ⁷ to establish the total 25-hydroxyvitamin D concentration defining vitamin D deficiency. Each of these effects, as a result of decreasing total 25-hydroxyvitamin D, has a different threshold concentration where the event accumulates and as a result the recommended concentration defining vitamin D deficiency varies from 30 nmol/L to 110 nmol/L. As an alternative approach increasing supplements of vitamin D and calcium were given until the total 25-hydroxyvitamin D concentration of 50 nmol/L established the point where no further decrease in PTH resulted. ⁸ Reviewing this heterogeneous data and recognizing that the majority of detrimental effects accumulate below 50 nmol/L, it was proposed ⁹ that total 25-hydroxyvitamin D concentrations between 25 nmol/L and 50 nmol/L be defined as vitamin D insufficient and below 25 nmol/L as vitamin D-deficient.

The clinical recommendations that have been made and the definitions produced for the various deficient states amount to an art form in terms of interpretation of the data. An additional complication that has been largely ignored is the significant differences that exist between the assays employed to measure total 25-hydroxyvitamin D. Several publications have highlighted problems of accuracy, demonstrating the lack of correlation between the assays that are available ^10,11 and also highlighting the problems of the immunoassays inability to measure 25-hydroxyvitamin D₂. ¹² As a result of the immunoassay problems that have been experienced, a significant number of laboratories have turned to alternative methods and in particular liquid chromatography tandem mass spectrometry (LC MS/MS) has gained in popularity. At first glance LC MS/MS would appear to offer several advantages over immunoassay including improved accuracy and better measurement of 25-hydroxyvitamin D₂, but the performance of these assays in the DEQAS scheme has not reflected this optimistic belief. Carter and Jones ¹ highlight the problems that exist with current LC MS/MS methods and the significant variability in results obtained by the LC MS/MS group within the DEQAS scheme. Obvious problems that exist with assay standardization include errors that can be introduced in the preparation of standards in different matrices, use of different internal standard material, alternative approaches to calibration and the lack of an International Standard material. It was clearly demonstrated that by using common standard materials with defined 25-hydroxyvitamin D₂/D₃ concentrations, provided by a commercial manufacturer, the between-laboratory variability could be significantly reduced and therefore consistency of results improved across the laboratories using different LC MS/MS methods. Such an improvement in assay performance would hopefully lead to a more consistent definition of vitamin D deficiency with better comparability between clinical trials using different outcomes to measure vitamin D deficiency. In addition, an improved outcome for patients receiving vitamin D supplementation should be seen as the target total 25-hydroxyvitamin D concentration achieved in such trials would have greater comparability between the studies.

Despite the welcome findings reported, and in particular the reduction in laboratory variability in total 25-hydroxyvitamin D measurement, several problems are highlighted that need to be urgently addressed by the scientific community. Although the results obtained in this study are more consistent this does not necessarily mean that they are correct and that the true circulating total 25-hydroxyvitamin D concentrations are being measured. The provenance of the standard material used in this study is unknown and the methods used to establish the concentration of the standard material have not been provided. It could be that uniformity of materials employed in-house and consistent methodology used to assign standard and calibrator values would result in similar improvements in the comparability of LC MS/MS methods. The results obtained using the commercial standard appear to be higher for D₃ and lower for D₂ compared with current values and the continual drift upward of measured total 25-hydroxyvitamin D concentrations in recent years is perhaps being reflected in the ever increasing concentrations recommended as the target to be achieved to ensure 25-hydroxyvitamin D repletion. Gas chromatography MS method may be established as the gold standard methodology for total 25-hydroxyvitamin D. Of prime importance is that an internationally agreed standard material for both 25-hydroxyvitamin D₂ and D₃ needs to be produced that can be utilized worldwide to improve not only LC MS/MS consistency but also immunoassay comparability.