Abstract
Follicular thyroid tumors pose a diagnostic challenge on the preoperative level, as the discrimination between follicular thyroid carcinoma (FTC) and adenoma (FTA) demands careful histopathological investigation. Moreover, prognostication of FTCs is mostly based on tumor size and extent of invasive properties, while immunohistochemical markers pinpointing high-risk cases are lacking. We have routinely established a Ki-67 labeling index for follicular thyroid tumors since 1999. To assess the potential value of Ki-67 as an adjunct tool to (1) correctly separate FTCs from FTAs and (2) help identify poor-prognosis FTCs, we collected histopathological and clinical data from 818 follicular thyroid tumors with a histological Ki-67 labeling index established in clinical routine practice (516 FTAs, 252 FTCs, and 50 follicular thyroid tumors of uncertain malignant potential (FT-UMPs)). The Ki-67 labeling index was higher in FTCs (mean 5.8%) than in FTAs (mean 2.6%) (P < 0.001), and a receiver operating characteristic curve analysis revealed a cut-off value of 4% to separate FTC from FTA with a sensitivity and specificity of 65% and 83%, respectively. Similarly, a Ki-67 labeling index above 4% was found to identify FTCs that later metastasized from clinically indolent FTCs with a sensitivity and specificity of 80% and 48%, respectively. Ki-67 constituted an independent predictor of future FTC metastases/recurrence and death of disease, and a value > 4% was a reliable prognostic marker within individual pT staging groups. We conclude that Ki-67 is a potentially valuable marker for the prognostication of FTCs, and future implementation in the histopathological assessments of follicular thyroid tumors could be beneficial if reproduced in international series.