Evaluation of immunoassays for the measurement of soluble transferrin receptor as an indirect biomarker of recombinant human erythropoietin misuse in sport

Soluble transferrin receptor (sTfR) has been proposed as an indirect biomarker of the misuse of recombinant human erythropoietin in sport. An extended validation of four commercially available immunoassays for its measurement in serum is presented. Two ELISA techniques (ELISA1: Orion Diagnostica; ELISA2: R&D Systems), an immunoturbidimetric technique (Turbid: Roche Diagnostics), and a nephelometric technique (Nephel: Dade Behring) were investigated. Intra-laboratory precision better than 3% and correct accuracies were obtained for the Turbid and Nephel techniques using autoanalysers. Slightly worse precision (but always better than 11%) and correct accuracies were also obtained in almost all cases for the two ELISA techniques. Inter-laboratory results showed higher concordances for the ELISA procedures (intraclass correlation coefficients of 0.848 for ELISA1 and 0.973 for ELISA2 which was clearly better). Inter-technique correlations were good for the four techniques with lower dispersions found for the techniques using autoanalysers, i.e. Turbid and Nephel. While Turbid and ELISA1 results (expressed in mg/l) were comparable, results obtained with Nephel were approximately 2.7 times lower. The relationship between those three techniques was maintained when compared with ELISA2, which uses different units (nmol/l). We conclude that ELISA2 and Nephel in our hands were the most suitable techniques in terms of sensitivity, precision and accuracy, and adequacy of the calibration curve for the measurement of sTfR in real serum samples. Discrepancies observed in the results obtained with the different sTfR techniques showed that different reference standards were used and harmonization is recommended in order to obtain comparable results.