Abstract

Background: The optimal duration of antithyroid drug (ATD) therapy in Graves' disease is uncertain. Although guidelines recommend 12-18 months, shorter courses may be sufficient in selected patients. We evaluated whether ATD duration <12 months is non-inferior to 12-18 months for relapse among patients with Graves' disease and lower thyrotropin receptor antibody (TRAb) levels (>1.8 and <10.0 U/L) at diagnosis.

Methods: We analyzed real-world data from adults with Graves' disease and baseline TRAb <10 U/L treated with ATDs at a single center. The primary endpoint was relapse within 12 months of ATD cessation. A prespecified risk-difference non-inferiority framework was used, with a primary margin of +5% and an exploratory margin of +10%. Treatment effects were estimated using unadjusted analyses, inverse-probability-of-treatment weighting (IPTW; n = 369), and 1:1 propensity-score matching (PSM; 104 matched pairs). Secondary outcomes included long-term relapse (median follow-up 58 months), TRAb levels at cessation, and restricted mean survival time (RMST) over 60 months.

Findings: At 12 months, relapse occurred in 23/134 (17.2%) patients treated <12 months and 48/235 (20.4%) treated for 12-18 months. The IPTW-adjusted risk difference was -1.5% (CI -11.0 to +8.0), meeting the 10% but not the 5% non-inferiority margin. The unadjusted estimate (-3.2%) met both margins, whereas the PSM estimate (+2.9%, CI -7.3 to +13.1) met neither because of reduced precision. Long-term outcomes were similar between groups (IPTW Cox HR 0.92 [CI 0.65-1.31]; RMST difference +3.2 months over 60 months [-2.1 to +8.4]). TRAb levels at cessation were comparable.

Conclusions: Among patients with Graves' disease and lower TRAb levels, shorter ATD courses (<12 months) produced outcomes broadly comparable to 12-18 months. Although strict non-inferiority was not confirmed at a 5% margin, findings meeting an exploratory 10% threshold support TRAb-guided individualization of ATD duration and justify prospective randomized evaluation.

Keywords: Graves; antithyroid drugs; disease; non-inferiority; relapse risk; treatment duration.