Graves' disease, the most common form of hyperthyroidism in children, is caused by Thyrotropin (TSH) Receptor Antibodies (TRAbs) that mimic the action of TSH. The disease leads to significant morbidity in children both due to the prolonged course of antithyroid medication often required for sustained immunological remission and the high risk of relapse when medication is withdrawn. The ability to predict which patients are most likely to fail medical management would greatly improve the choice of therapy. In the past, large goiter size, age at diagnosis, increased biochemical severity, and decreased body mass index have all been associated with a poorer prognosis, but these clinical indicators lack sensitivity and specificity. Preliminary data suggest that the new TRAb assays are both sensitive and specific for the measurement of TRAbs in children with Graves' disease. In addition, variation in these antibodies over time is not the same in all patients. The goal of this proposal will be to prospectively follow children with newly diagnosed Graves' disease and use microarray technology to determine if there are genes whose expression differ in patients who respond to medical therapy versus those who will need more definitive therapy earlier in their disease.