Predicting radioiodine therapy response in advanced and recurrent thyroid cancer.

Radioiodine therapy of differentiated thyroid cancer is effective and possibly curative in patients with metastatic lesions that retain iodine avidity. However, in clinical practice, iodine avidity in any metastatic site is usually unknown at the point of initial treatment. Avidity prediction could enable more tailored initial radioiodine treatment. This work aimed to establish the best predictive factors for iodine avidity and subsequent treatment response. Thirty-one patients with metastatic or recurrent thyroid cancer sites in which iodine avidity could be assessed by image-based analysis were included. In tissue specimens from primary surgery, immunohistochemical expression of thyroglobulin (Tg), thyroid peroxidase (TPO), sodium-iodide symporter (NIS), and Ki67 and mutational status of BRAF, RAS, and the TERT promoter were analyzed. Biochemical and structural responses were assessed by serum Tg response and radiological evaluation. High-risk histology (widely invasive follicular, tall cell subtype papillary, differentiated high-grade, or poorly differentiated thyroid carcinoma) combined with the expression of TPO, Tg, and cytoplasmic NIS performed well in predicting treatment response and iodine avidity. Approximately half of the variation between patients was explained by those variables, outperforming pT stage and mutational status. TPO and Tg expression were the largest contributors to prediction modeling. Radioiodine treatment response and iodine avidity in thyroid cancer can be predicted from histopathological analysis of the primary tumor. High-risk histology and the expression of TPO, Tg, and NIS were robust and informative predictors. This may be used to adapt treatment strategy in adjuvant and metastatic therapy settings.