Soluble transferrin receptor (sTfR) concentration quantified using two sTfR kits: analytical and clinical performance characteristics

Abstract

We compared the analytical and clinical performance characteristics of the Ramco and R&D Systems enzyme-linked immunosorbent assays (ELISAs) for quantifying serum levels of soluble transferrin receptor (sTfR). In addition, we determined both the number of samples required to determine the true individual mean sTfR concentration for a single individual and the critical difference (CD) between serial measurements that indicates a statistically significant change in sTfR concentration. sTfR concentration was determined in 127 serum samples selected retrospectively from males (n=32) and non-pregnant (n=40) and pregnant women (n=55). Intra- and inter-assay precision for both methods was good (CV values 5–10%) to excellent (CV values <5%) over a wide range of sTfR concentrations. Correlation between these methods was good (r=0.93); however, sTfR values by the R&D kit were ∼2.9 times higher than values obtained using the Ramco kit on the same serum samples. Nevertheless, receiver–operator characteristic (ROC) curve analysis demonstrated that the diagnostic accuracy of both assays in discriminating between patients with iron-deficiency anemia (IDA) or anemia of chronic disease (ACD) was high (area-under-the-curve (AUC) values >0.95) and not significantly different (P=0.480). We determined that a minimum of 8 samples are required to determine an individual’s true sTfR concentration, while a >40% difference between serial sTfR measurements would be required to indicate a statistically significant change in sTfR concentration. We concluded that both the Ramco and R&D Systems sTfR methods have similar analytical and clinical performance characteristics and were likely to be equally useful in discriminating between patients with biochemically defined IDA or ACD.

Introduction

The cellular transferrin receptor (TfR) molecule is a transmembrane protein consisting of two 760-amino acid glycoprotein monomers linked by two disulfide bridges to form a dimer of approximately 190 kilodaltons (kDa) (Fig. 1). As the name suggests, TfR binds transferrin, the principal iron transport protein found in the blood [1], [2]. In 1986, Kohgo [3] was the first to demonstrate the presence of a soluble form of TfR (sTfR) in human serum that was identified subsequently as a truncated form of cellular TfR derived from proteolytic cleavage of the extracellular segment [4], [5], [6]. TfR is found in highest concentration on the surface of cells requiring large amounts of iron, such as hemoglobin-synthesizing cells of the reticuloendothelial system (RES) (i.e., bone marrow, liver, and spleen) [7] and the placenta [8], [9], [10], [11].

The principal clinical utility of sTfR measurement has been shown to be in the assessment of total erythropoiesis [12], [13], [14], in diagnosing ‘functional iron deficiency’ [15], [16], [17], in distinguishing iron deficiency anemia (IDA) from anemia of chronic disease (ACD) [18], [19], in the overall evaluation of iron status [20], [21], in the detection of athletes suspected of ‘doping’ with erythropoietin [22], [23], in monitoring recovery of erythropoietic activity after bone marrow transplantation [1], and in quantifying the response to therapy with human recombinant erythropoietin [24], [25], [26]. Because serial sTfR measurements are used often in the aforementioned cases, it is of interest to know both the number (N) of serum samples required to achieve an estimate of the sTfR concentration that is within 10% accuracy of the true concentration and the critical difference (CD) between sTfR concentrations, obtained on consecutive patients’ samples, that indicates a significant change has occurred beyond the maximum expected change due to the combined and inescapable effects of analytical and intra-individual biological variation on sTfR results.

We evaluated the analytical and clinical performance characteristics of two commercially available kits for the measurement of serum sTfR concentration; the number of sTfR samples required to achieve 10% accuracy compared to the true sTfR concentration; and, the CD indicative of a significant change in sTfR concentration between consecutive serum samples.