Thyroid-stimulating antibody assay using a human thyrotropin receptor transfected cell line: relationship to clinical features of graves' disease.

To evaluate a Chinese hamster ovary cell line transfected with the human thyrotropin (thyroid-stimulating hormone or TSH) receptor (CHO-TSHR) for its use in routine clinical testing and compare it with the currently used Fisher rat thyroid cell line (FRTL-5) relative to the cyclic adenosine monophosphate (cAMP) responses to thyroid-stimulating antibodies (TSAb). We analyzed 112 serum samples prospectively. TSAb were measured concurrently by both CHO-TSHR cells and FRTL-5 cells. In addition, thyrotropin-binding inhibitory immunoglobulin (TBII) was measured by a radioreceptor assay. Among the study subjects, 51 had active Graves' disease, 40 had other thyroid disease, and 21 had no thyroid disease. Of 51 patients with Graves' disease, 38 (74%) had positive results in both cell assays, 4 (8%) had negative results in both assays, and 9 (18%) had positive responses with CHO-TSHR cells but negative results with FRTL-5 cells. No patient had negative findings with CHO-TSHR cells and positive results by FRTL-5 cells. All 61 patients in the other two groups had negative results by both assays (diagnostic specificity = 100%). The diagnostic sensitivity and accuracy of TSAb activity by the two cell lines (CHO-TSHR versus FRTL-5) were 92.2% versus 74.5% and 96.4% versus 88.4%, respectively. A significant correlation was noted between the TSAb activities in active Graves' sera by the two assays (r = 0.58; P<0.0001); however, values by the CHO-TSHR method were significantly higher (P<0.003). TSAb activities determined with both cell lines were positively correlated with the 24-hour thyroidal 131 I uptake (r = 0.62 and P<0.0005 in CHO-TSHR cells; r = 0.41 and P<0.05 in FRTL-5 cells). The difference in TSAb activity between patients with ophthalmopathy and those without ophthalmopathy was statistically significant with both FRTL-5 (P<0.01) and CHO-TSHR (P = 0.05) assays. The correlation between the severity of eye disease and TSAb activities was significant only with the FRTL-5 assay (P = 0.009) but not with the CHO-TSHR assay (P = 0.16). TSAb activity showed a significant positive correlation with serum thyroxine, triiodothyronine, and TBII by both cell assays. Neither cell line showed a significant correlation between TSAb activity and antimicrosomal antibody titer. The CHO-TSHR cell line is superior to the standard FRTL-5 cell line in the cAMP bioassay for TSAb with reference to diagnostic sensitivity and accuracy, correlation with disease activity, simplicity, and cost.