LAUSR

Abstract Context The conventional treatment of nonmedullary thyroid carcinoma (NMTC) includes surgical resection, thyrotropin (TSH) suppression, and 131-iodine. Some patients develop persistent/recurrent metastatic disease requiring expensive alternative therapies, such as external radiation and multikinase inhibitors, which may have clinically significant side effects. Recent in vitro studies, in vivo studies in animals, and association studies in humans suggest that metformin, an inexpensive medication with a modest side effect profile, may help prevent or treat NMTC. No interventional trials analyzing the effect of metformin have been performed in humans. Objective We hypothesize that metformin administration will decrease serum thyroglobulin concentration (Tg), a surrogate marker for NMTC burden. Methods This retrospective institutional review board–approved study included 10 patients with persistent/recurrent NMTC who had exhausted conventional therapies including total thyroidectomy and 131-iodine. Five had detectable disease on computed tomography imaging. All had biochemical evidence of NMTC with Tg > 2.0 ng/mL with nondetectable serum thyroglobulin antibody concentrations. Five elected to have metformin treatment at doses varying from 500 to 2000 mg/day for 2 to 5 months. The remaining 5 served as untreated controls. Statistical significance was determined by the Mann–Whitney test. Results Tg decreased (mean decrease = 21.7 ± 8.4%) in all 5 patients receiving metformin and increased (mean increase = 16.6 ± 12.1%) in all 5 controls (P < .01). TSH did not change significantly in either group. Conclusion In summary, metformin caused a TSH-independent Tg decrease in patients with persistent/recurrent NMTC. More extensive studies are required to determine if metformin slows NMTC progression