Evaluation of an automated soluble transferrin receptor (sTfR) assay on the Roche Hitachi analyzer and its comparison to two ELISA assays

BackgroundSoluble transferrin receptor (sTfR) assays are currently not standardized. This hinders data comparison between studies and also affects the use of a recently proposed model to estimate body iron.MethodsWe evaluated the analytical performance of a fully automated sTfR immunoturbidimetric assay (Roche Diagnostics) and compared it with two ELISA assays (Ramco Laboratories and an in-house ELISA assay used in the body iron model).ResultsThe Roche assay showed excellent intra- and inter-assay precision (CV < 5%). Prolonged exposure of serum samples to room temperature and multiple freeze–thaw cycles did not affect sTfR concentrations. Receiver–operator characteristic curve analysis demonstrated that the Roche assay (area-under-the-curve (AUC) = 0.882) was superior to the Ramco assay (AUC = 0.794) in predicting iron deficiency (defined as serum ferritin < 10 μg/L; P = 0.013). Method comparison between the Roche and the two ELISA assays showed good correlations (r > 0.8); however, sTfR values by the Roche assay were on average 30% lower than values obtained with the two ELISA assays.ConclusionssTfR data measured with an immunoturbidimetric assay can be compared to a commonly used ELISA assay, and can be used in the body iron model through regression equations obtained in the present study.