Toward Personalized Risk Stratification in Advanced Follicular Thyroid Carcinoma.

Abstract

BackgroundAdvanced follicular thyroid carcinoma (FTC) represents a clinically heterogeneous subgroup of differentiated thyroid cancer with variable long-term outcomes. We aimed to develop and internally validate a population-based nomogram for predicting 3-, 5-, and 10-year disease-specific survival (DSS) in patients with stage III-IV FTC using SEER registry data.Materials and MethodsThis retrospective cohort study utilized data from the SEER 17 registry (2010-2022). A total of 1027 patients with stage III-IV FTC were included and randomly assigned to training (n = 718) and validation (n = 309) cohorts. Clinical and demographic variables were analyzed. Univariable and multivariable Cox proportional hazards regression analyses were performed to identify independent prognostic factors. A nomogram was constructed based on the final multivariable model. Predictive performance was assessed using ROC and calibration analyses.ResultsThe mean age of the cohort was 63.5 years (±11.7), and 59.3% of patients were female. Multivariable analysis identified age, stage IV disease, N1 nodal status, unmarried status, and surgical treatment category as independent predictors of DSS. The nomogram demonstrated strong discriminative performance, with AUC values of 0.953, 0.966, and 0.903 for predicting 3-, 5-, and 10-year DSS, respectively. Calibration analyses showed acceptable agreement.ConclusionThis nomogram provides a population-level estimate of DSS in advanced FTC. While the model may support preliminary risk stratification using readily available registry variables, its clinical interpretation should consider the absence of key biologic and treatment-response factors not captured in SEER, including vascular invasion, molecular alterations, and systemic therapy data.