Predictive Performance of 2 Immunoassays in Patients with Graves Disease Undergoing Radioiodine Therapy: Prospective Study.

Context: Radioactive iodine (RAI) treatment is a well-established successful treatment for Graves disease (GD) but causes an increment in autoantibodies, particularly TSH receptor antibodies.

Objective: To assess the performance and prognostic role of Immulite thyroid-stimulating immunoglobulin (TSI) and Elecsys thyrotropin receptor antibodies (TRAbs) immunoassays in patients with GD undergoing RAI therapy.

Methods: Clinical and laboratory data of 188 patients (156 GD and 32 toxic nodule goiter), undergoing RAI therapy between January 2018 and January 2022 were prospectively collected over 12 months. Refractoriness was defined as persistent or recurrent hyperthyroidism 6 months post-RAI treatment without levothyroxine. Statistical analysis included descriptive statistics, logistic regression, and generalized estimated equations.

Results: Patients had a mean age of 46.4 years, and 78.2% were women. RAI therapy was indicated in 94.2% due to uncontrolled hyperthyroidism or ATD therapy relapse (median of 35 months). Immulite TSI showed higher clinical sensitivity and accuracy (area under the curve [AUC]: 0.98, sensitivity 92.0%, accuracy 98.4%) than Elecsys TRAb (AUC: 0.97, sensitivity 82.1%, accuracy 91.2%). Successful treatment was achieved in 112 of 126 GD patients (89%). Thyroid volume, 2-hour iodine-131 uptake, free thyroxine and thyroxine levels, Elecsys TRAb, and Immulite TSI were significantly higher in the refractory group (P < 0.05), despite most patients receiving >300 Gy of RAI. Longitudinal thyrotropin evaluation predicted treatment response at 12 months (P = .01), whereas autoantibodies did not.

Conclusion: Refractoriness to RAI therapy was associated with higher levels of Immulite TSI and Elecsys TRAb prior to treatment. Although AUCs for both assays were equivalent, Immulite TSI demonstrated superior clinical sensitivity and accuracy. Despite distinct autoantibody patterns emerging post-RAI, longitudinal monitoring did not predict treatment response after 1 year but indicated persistently high concentrations.