ABSTRACT
A new procedure for measuring patient serum thyrotropin receptor (TSHR) autoantibodies is described in which the autoantibodies inhibit binding of a human monoclonal thyroid stimulating antibody M22 (labeled with biotin) to TSHR-coated enzyme-linked immunosorbent assay (ELISA) plate wells. In the assay, M22-biotin binding is detected by addition of streptavidin peroxidase. The M22 based assay was more sensitive than a similar ELISA based on inhibition of TSH-biotin binding to TSHR coated wells with 1 U/L of NIBSC 90/672 giving approximately 35% inhibition in the M22-based system compared to approximately 15% inhibition in the TSH-based ELISA. This had an important impact on the precision of the 2 assays with the M22-based ELISA showing an interassay coefficient of variation (CV) of 10% at 1 U/L whereas the TSH-based ELISA had an interassay CV of 20% at 1 U/L. Analysis of sera from 307 control subjects without a diagnosis of Graves' disease indicated that only 2 (0.65%) gave inhibition of M22 binding values of greater than 10% (11% and 12% inhibition). In the case of sera from 108 patients with Graves' disease (treated and untreated) 103 (95%) gave inhibition of M22 binding values of 14% or greater. Receiver operating characteristic (ROC) plot analysis showed that 100% specificity for TSHR autoantibody detection in Graves' disease was obtained at 95% sensitivity for the M22-based ELISA and 89% sensitivity for the TSH-biotin-based ELISA. Inhibition of M22 binding to the TSHR was closely correlated to inhibition of TSH binding in the 108 Graves' sera (r = 0.99). However, inhibition of M22 binding was almost always greater resulting in improved sensitivity and precision.
