Early predictors of 6-month thyroid-stimulating antibody levels after radioactive iodine therapy for Graves' disease: implications for pregnancy planning.

Current guidelines recommend a 6-month contraception period following radioactive iodine therapy (RAIT) for Graves' disease (GD), based primarily on radiation safety considerations; however, they do not account for the post-treatment surge in thyroid-stimulating antibodies (TSAb), which poses risks for fetal and neonatal GD. This retrospective study aimed to identify factors influencing TSAb trajectories and construct a predictive model for TSAb levels at 6 months after RAIT. We analyzed 82 patients who underwent RAIT, with TSAb values standardized to the upper limit of normal (ULN) to account for different assay methods. Multivariable linear regression identified the duration of antithyroid drug and potassium iodide administration, pre-RAIT TSAb levels, changes in TSAb at 3 months (3M), fractional reduction in TW at 3M, and assay method as significant predictors of TSAb levels at 6 months (adjusted R2 = 0.87). A prediction score derived from this model demonstrated excellent diagnostic performance. Receiver operating characteristic analysis yielded an area under the curve of 0.99 for predicting TSAb >3.3 ULN, corresponding to the threshold for fetal GD monitoring at approximately 20 weeks of gestation, and 0.94 for TSAb >1.0 ULN. The prediction score significantly outperformed single predictors, including baseline TSAb levels and early post-TSAb changes. In conclusion, we developed a highly accurate predictive model for TSAb levels at 6 months after RAIT by integrating early post-treatment data with pre-RAIT clinical parameters. This model enables identification of patients at risk of persistent TSAb elevation, supporting individualized counseling regarding pregnancy planning beyond the conventional 6-month contraception period.