Abstract
BACKGRUOUND: To determine whether baseline thyroid-stimulating immunoglobulin (TSI) bioassay or its early response upon treatment with an anti-thyroid drug (ATD) can predict prognosis of Graves' disease (GD) in real-world practice. METHODS: This retrospective study enrolled GD patients who had previous ATD treatment with TSI bioassay checked at baseline and at follow-up from April 2010 to November 2019 in one referral hospital. The study population were divided into two groups: patients who experienced relapse or continued ATD (relapse/persistence), and patients who experienced no relapse after ATD discontinuation (remission). The slope and area under the curve at 1st year (AUC1yr) of thyroid-stimulating hormone receptor antibodies including TSI bioassay and thyrotropin-binding inhibitory immunoglobulin (TBII) were calculated as differences between baseline and second values divided by time duration (year). RESULTS: Among enrolled 156 study subjects, 74 (47.4%) had relapse/persistence. Baseline TSI bioassay values did not show significant differences between the two groups. However, the relapse/persistence group showed less decremental TSI bioassay in response to ATD than the remission group (-84.7 [TSI slope, -198.2 to 8.2] vs. -120.1 [TSI slope, -204.4 to -45.9], P=0.026), whereas the TBII slope was not significantly different between the two groups. The relapse/persistence group showed higher AUC1yr of TSI bioassay and TBII in the 1st year during ATD treatment than the remission group (AUC1yr for TSI bioassay, P=0.0125; AUC1yr for TBII, P =0.001). CONCLUSION: Early changes in TSI bioassay can better predict prognosis of GD than TBII. Measurement of TSI bioassay at beginning and follow-up could help predict GD prognosis.