RELIABILITY OF PLASMA TESTOSTERONE ASSAYS BY COMPETITIVE PROTEIN BINDING METHODS

Abstract

ABSTRACT This report examines the reliability of different methods for competitive protein binding assay of plasma testosterone. Most of the methods give results similar to those reported by the most reliable of the competitive protein binding methods and by established double isotope derivative methods. In a few instances poor reliability was demonstrated. However, the reliability of the remaining methods usually had not been investigated thoroughly. An adequate examination of accuracy was one of the most frequent omissions in the investigation of the reliability of these methods. Differences between direct standards and samples containing known amounts of testosterone purified in the same manner as unknown samples can develop easily in competitive protein binding methods. These differences may be caused by substances other than the steroid being measured or by alteration of the labelled or unlabelled steroid. The interfering substance or altered steroid may be present before or appear during or after purification. The impurity may or may not be removed in purification and it may or may not bind with the binding protein. To detect these types of errors, accuracy studies should be done throughout the range of concentrations in which it is expected to apply the method.