Immunoradiometric Assay for Serum Thyroglobulin: Semiquantitative Measurement of Thyroglobulin in Antithyroglobulin-Positive Sera*

Abstract

All RIAs for human serum thyroglobulin (Tg) described hitherto are based on the same principle; namely, competitive binding between cold and labeled antigen, followed by separation of bound from free antigen by precipitation of the former by a second antibody. These RIAs for Tg are accurate only when applied to sera free of detectable anti-Tg because the presence of the latter can result in either falsely elevated or falsely depressed values. This report describes a solid phase, sandwich-type, immunoradiometric assay (IRA) for serum Tg. Tg in the sample or standard is first bound to plastic cups coated with rabbit anti-Tg and then measured by the binding of rabbit [125I]anti-Tg. The sensitivity of this assay (detection limit, 2.5 ng Tg/ml serum) and its reproducibility, as indicated by intraassay coefficients of variation(CV20 ng, 11.3%; CV115 ng, 4.1%; CV310 ng, 7.0%) and interassay coefficeients of variation (CV20 ng, 14.0%; CV115 ng, 7.2%; CV310 ng, 7.1%)are comparable to or better than those previously described. Normal values are <40 ng Tg/ml. The major advantage of this method is that Tg can be determined at least semiquantitatively in sera with abnormal anti-Tg concentrations. The presence of anti-Tg gives rise to depressed values for Tg, i.e. Tg measured < Tg actually present, because Tg in Tg-anti-Tg complexes is measured with lower sensitivity than free Tg. However, the interference by anti-Tg is unidirectional, i.e. falsely elevated Tg values do not occur and falsely depressed Tg values can be corrected by means of Tg recovery studies. Recovery studies are particularly indicated in anti-Tg-positive sera with borderline Tg values (Tgfound, 10–40)ng/ml).