Counterpoint: legitimate and illegitimate tests of free-analyte assay function: we need to identify the factors that influence free-analyte assay results.

There have been, and continue to be, reports of discordant free thyroxine (T4) 3 measurements when different free-T4 methods are applied to the same serum samples (1– 4 ). Our laboratory has attempted to identify some of the factors that contribute to these disparities. There are 4 separate components of serum T4: free T4, albumin-bound T4, transthyretin-bound T4, and thyroxine-binding globulin (TBG)-bound T4. Free T4 is the one component that moves across semipermeable membranes during equilibrium dialysis and ultrafiltration. As Midgley and Christofides point out, the gold standard free-T4 methods are designed to measure this component. We applied analog-type methods and an equilibrium-dialysis/immunoassay gold standard method to carefully prepared solutions in which we varied these 4 components of serum T4. This equilibrium-dialysis method agrees closely with a liquid chromatography– tandem mass spectrometry method applied to serum ultrafiltrates (4 ). We also applied this equilibrium dialysis/immunoassay and a group of 7 analog-type free-T4 assays to solutions of T4 that did not include T4-binding proteins (5 ); that is, the only form of T4 was free T4. The analog-type assays required 28to 588-fold higher concentrations of free T4 than the equilibrium dialysis/immunoassay to generate similar assay results. These results highlight the widely differing characteristics of the antibodies used in these assays. Midgley and Christofides state that analog-type methods are not suitable for application to “pure” analyte solutions. We understand that they are not suitable for the measurement of free T4 in the absence of protein-bound T4; however, we are not persuaded that applying any assay to the substance it reports is inappropriate, when the goal is to identify the factors that influence assay results. When albumin, transthyretin, and TBG are held constant and the total T4 concentration is varied, the concentrations of free T4, albumin-bound T4, transthyretin-bound T4, TBG-bound T4, and total T4 are aligned with one another. That is, they are proportionate covariables. In a recent experiment, the total T4 concentration was varied while albumin, transthyretin, and TBG concentrations were within the reference interval and constant. Seven analog-type free-T4 assays were applied to these T4 solutions. Analog-type free-T4 estimates correlated closely with dialyzable T4 concentrations (Fig. 1). This result prompts a question: What happens to analog-type free-T4 estimates when these 5 T4 components are not aligned? In a previous study, albumin, transthyretin, TBG, and total-T4 concentrations were proportionate covariables, and dialyzable T4 was held constant (6 ). Five analog-type free-T4 assays were applied. Free-T4 estimates obtained with analog-type assays reflected the concentrations of T4-binding proteins and total T4. They did not reflect the fixed dialyzable T4 concentration. These data are direct evidence that total T4 and/or T4-binding proteins influence such assay results. In another study, total T4 was varied in the presence and absence of a T4-binding protein (7 ). Individual T4-binding proteins were added at clinical concentrations to the same concentrations of total T4. Four analog-type free-T4 assays were applied. Similar free-T4 estimates were obtained with similar concentrations of total T4 in both the absence and the presence of a T4-binding protein. Analog-type free-T4 estimates did not follow dialyzable T4 concentrations. They were more nearly related to the total T4 concentration. In yet another study, an analog-type assay was applied to a series of solutions in which free T4 progres1 Department of Biochemistry and 2 Departments of Internal Medicine and Pathology, Loma Linda University School of Medicine, Loma Linda, CA. * Address correspondence to this author at: Department of Biochemistry, Loma Linda University, Loma Linda, CA 92350. Fax 909-558-4887; e-mail [email protected]. Received November 20, 2008; accepted December 23, 2008. Previously published online at DOI: 10.1373/clinchem.2008.120154 3 Nonstandard abbreviations: T4, thyroxine; TBG, thyroxine-binding globulin. Clinical Chemistry 55:3 442–444 (2009) Point/Counterpoint