Abstract
It is well established that patients with the most differentiated thyroid cancers have a good prognosis, whereas when the disease develops into radio-iodine refractory thyroid cancer (RAIR) their prognosis is poor and the 10-year survival rate is low. At present, the therapeutic methods for RAIR are limited and have low efficacy. As a consequence, several models have been developed for predicting RAIR. The aim of this review was to describe recent developments regarding the factors that influence and predict the occurrence of RAIR. Many variables including demographic characteristics, tumor clinicopathology, serology changes, disease characteristics, and PET/CT results have been shown to be independent factors that influence the development of RAIR. The cut-off value derived from multivariate prediction models therefore effectively predicts the onset of RAIR. However, the current models for predicting RAIR were obtained through retrospective studies, and the prospective prediction studies are needed in the future to confirm their validity.