Evaluation of the analytical and clinical performance of the Fujirebio Lumipulse® G 25-OH vitamin D assay

Abstract

Recently, Fujirebio introduced the Lumipulse G 25-OH vitamin D assay, the first non-competitive sandwich assay that uses a monoclonal antibody against a hapten-antibody immunocomplex. It is a two-step fully automated immunoassay without additional pre-treatment steps. In the first step, 25(OH)D is separated from vitamin D-binding protein (VDBP) and binds to a monoclonal sheep anti-25(OH)D antibody coated on solid phase particles. In the second step, this complex is bound by a second monoclonal antibody labelled with alkaline phosphatase.¹ According to the manufacturer, this new assay has an improved analytical performance. The two-step assay format should allow a more efficient separation of 25(OH)D from its carriers VDBP, albumin and lipoproteins. Furthermore, it can be expected that interferences, such as heterophile antibodies, are eliminated more effectively.

The present study evaluates the analytical performance of the Fujirebio 25(OH)D assay (FUJIREBIO Inc., Tokyo, Japan) on a LUMIPULSE® G600 analyzer (FUJIREBIO Inc., Tokyo, Japan) in comparison to a commercial liquid chromatography-tandem mass spectrometry (LC-MS/MS) method. Analytical performance was evaluated by the following components: estimation of limit of blank (LoB), limit of detection (LoD), limit of quantitation (LoQ),² linear range, recovery and imprecision. Accuracy was analysed in 455 unselected residual serum samples from our routine work and in particular patient groups that are prone to interfere with immunoassays: pregnant women, renal and intensive care patients. In these patients, an altered VDBP concentration may influence the dissociation kinetics of 25(OH)D from its carriers, so that small amounts remain protein bound and are not measured by the assay. This results in underestimation of 25(OH)D. All samples were from individuals >18 years. Measurement was performed with the Lumipulse® G 25-OH vitamin D assay and a commercial NIST972a traceable LC-MS/MS method (RECIPE, Munich, Germany). Samples were tested with both methods under identical conditions without freezing or prolonged storage periods. The local ethics committee has approved the study.

In our hands, the Fujirebio Lumipulse 25(OH)D immunoassay showed performance characteristics comparable to those provided by Fujirebio. From repeated analyses of blank samples and samples with a low 25(OH)D concentration (mean: 1.31 ng/mL, SD: 0.32), we calculated an LoB of 1.09 ng/mL, an LoD of 1.6 ng/mL and an LoQ of 4.2 ng/mL. Serial 1:1 dilutions of a sample with a 25(OH)D concentration of 180 ng/mL revealed a linear range between 4 and 150 ng/mL (results of serially diluted samples: 90, 50, 27, 14.5, 7.6, 5.5 ng/mL). Recovery was tested mixing a high (76 ng/mL) and a low (8.7 ng/mL) sample at different proportions. The results show a slight over-recovery that was within acceptable limits (80–120%). The Passing-Bablock analysis for unselected samples and subgroups is presented in Figure 1 as well as the Bland-Altman plot in unselected samples.³ In unselected samples, the Fujirebio Lumipulse 25(OH)D immunoassay shows significant proportional bias of 11%. In selected subgroups, such as pregnant women, bias can reach up to 35% when compared with LC-MS/MS. The present results are in agreement with a previous assay comparison study by Cavalier et al.⁴ that showed a significant proportional bias of the Fujirebio Lumipulse 25(OH)D assay of −9% in unselected patients, −22% in pregnant women and −8% in renal patients. The study by Cavalier et al. includes additional subgroups, such as healthy Africans and osteoporotic patients. In these two groups, the Fujirebio Lumipulse 25(OH)D immunoassay displayed a significant constant bias of 4.2 ng/mL and −2.5 ng/mL, respectively. Although the Fujirebio assay shows systematic bias, the rather small dispersion of the results around the regression line suggests that the assay deals better with problematic samples. The visual impression of a low scatter around the regression line is supported by a concordance correlation coefficient (CCC) of 0.96 and a Pearson’s P of 0.98, which is higher than reported for other immunoassays.^5,6 Finally, intra- and inter-assay imprecisions were below 6.5% and 6.8%, respectively, at 25(OH)D concentrations between 5.8 and 91.7 ng/mL.

Figure 1.

PassingBablock analysis and BlandAltman plot in unselected sample (A)–(B). Passing-Bablock analysis in different subgroups (C)–(F).

In conclusion, the Fujirebio Lumipulse 25(OH)D assay appears to deal better than other automated immunoassays with common interferences, such as renal impairment or haemodialysis. This is probably the result of the two-step approach. However, the assay consistently underestimates 25(OH)D between 11 and 35%, depending on the patients tested. Fujirebio should address the systematic bias of their assay.

Footnotes

Acknowledgements

We would like to thank Dr. Vucetic Zivjena for her assistance in this research.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This research was funded by Fujirebio.

Ethical approval

The ethics committee of the Azienda Sanitaria dell’Alto Adige (ASDAA) has approved this study (REC nr.89-2015, 16 September 2015).

Guarantor

HM.

Contributorship

HM conceived the study, was involved in protocol development and edited the manuscript.

GS and LM were involved in patient recruitment and data analysis. GS wrote the first draft of the manuscript.

CD gained ethical approval.

All authors reviewed and edited the manuscript and approved the final version of the manuscript.

References

Lumipulse® 25-OH Vitamin D (package insert). Oct 2014 (ver.2) KEY-CODE: FRI94363.

Armbruster

Pry

. Limit of blank, limit of detection and limit of quantitation. Clin Biochem Rev 2008; 29: S49–S52.

Bland

Altman

. Comparing methods of measurement: why plotting difference against standard method is misleading. Lancet 1995; 346: 1085–1087.

Cavalier

Lukas

Bekaert

et al.

Analytical and clinical evaluation of the Fujirebio Lumipulse®G non-competitive assay for 25(OH)D and three immunoassays for 25(OH)D in healthy subjects, osteoporotic patients, third trimester pregnant women, healthy African subjects, hemodialyzed and intensive care patients. Clin Chem Lab 2016; 54: 1347–1355.

Cavalier

Lukas

Crine

et al.

Evaluation of automated immunoassays for 25(OH)-vitamin D determination in different critical populations before and after standardization of the assays. Clin Chim Acta 2014; 20: 431–460.

Farrell

Martin

McWhinney

et al.

State-of-the-art vitamin d assays: a comparison of automated immunoassays with liquid chromatography–tandem mass spectrometry methods. Clin Chem 2012; 58: 531–542.