Prediction of human iron bioavailability using rapid c-ELISAs for human plasma hepcidin

Item does not contain fulltext
BACKGROUND: Hepcidin is the central systemic regulator of iron metabolism, but its quantification in biological fluids is challenging. Rapid, accurate and user-friendly methods are needed. Our aim was to assess the ability of hepcidin as measured by three different c-ELISA assays to predict iron bioavailability in humans. METHODS: The three assays used were commercially available DRG and Peninsula assays and the c-ELISA method performed at Radboud University Medical Centre, Nijmegen, The Netherlands (Hepcidinanalysis.com), validated by comparative measurements with time-of-flight mass spectrometry. We analyzed plasma samples (n=37) selected to represent a broad range of hepcidin concentrations from a subgroup of healthy, iron-depleted women in a study assessing fractional absorption from iron supplements. RESULTS: In single regressions, all three c-ELISA assays were predictors of fractional iron absorption: R2=0.363 (DRG), R2=0.281 (Peninsula) and R2=0.327 (Hepcidinanalysis.com). In multiple regressions, models including hepcidin measured with either DRG-, Peninsula or Hepcidinanalysis.com explained 55.7%, 44.5% and 52.5% of variance in fractional absorption, and hepcidin was a strong predictor of fractional absorption irrespective of the hepcidin assays used. However, we found significant differences in absolute values for hepcidin between different methods. Both the DRG assay's (y=0.61x+0.87; R2=0.873) and the Peninsula assay's measurements (y=1.88x+0.62; R2=0.770) were correlated with Hepcidinanalysis.com. CONCLUSIONS: The biological variability in plasma hepcidin, (inter-sample CV) was 5-10-fold higher for both the Peninsula and DRG assay than the analytical variably (inter-run within-sample CV) suggesting substantial discriminatory power to distinguish biological hepcidin variation. Between methods, prediction of iron bioavailability in generally healthy iron depleted subjects appears comparable.