Validation of the International Medullary Thyroid Carcinoma Grading System in a Multicenter Cohort: Digital Image Analysis-Based Evaluation of Ki-67 Cutoffs and Assessment Methods

Abstract Although the International Medullary Thyroid Carcinoma Grading System (IMTCGS) was proposed to standardize the histologic grading of medullary thyroid carcinoma (MTC), the validity of its 5% Ki-67 index cutoff and the optimal assessment method remain to be fully established. This multicenter study aimed to validate the Ki-67 index component of the IMTCGS in a Korean cohort using digital image analysis (DIA) and to determine the most clinically relevant evaluation method. We retrospectively analyzed 187 surgically resected MTC cases across 16 institutions. Ki-67 indices were evaluated via DIA using two approaches: the hotspot method (highest proliferative area) and the average method (across three representative regions). Prognostic values of various Ki-67 cutoffs (5%, 3%, and ROC curve-derived thresholds) were evaluated independently and within the IMTCGS framework. IMTCGS high-grade status incorporating the 5% hotspot Ki-67 index emerged as an independent prognostic factor for recurrence-free survival (RFS), along with male sex, elevated postoperative serum calcitonin levels, and pN1b status. IMTCGS grading using the 5% average Ki-67 index failed to retain independent significance. While the composite high-grade IMTCGS classification remained an independent predictor of RFS in multivariate analysis, none of its constituent components (tumor necrosis, mitotic count, or Ki-67 index) retained independent prognostic significance when evaluated separately. This multicenter study validates the prognostic utility of the IMTCGS in Korean MTC patients. DIA-based hotspot Ki-67 assessment provides superior prognostic stratification compared to the average method, supporting its preference for routine clinical IMTCGS implementation.