Abstract
BACKGROUND: Primary thyroid lymphoma (PTL) is a rare and aggressive malignancy with a need for more precise prognostication tools due to the limitations of existing staging systems. This study aims to develop a nomogram to predict overall survival (OS) rates in PTL patients, addressing the gap in personalized treatment protocols. METHODS: We analyzed 1469 PTL cases. Cox regression analyses were used to identify key prognostic factors and construct a survival prognostic nomogram. The nomogram's performance was evaluated using receiver operating characteristic (ROC) curves and decision curve analysis. Additionally, a web-based dynamic nomogram was developed to estimate mortality risk for PTL patients. RESULTS: The nomogram exhibited high clinical utility and precision as determined by decision curve analysis and ROC curves. Furthermore, a novel risk stratification system was introduced. Kaplan-Meier survival curves illustrated significant differences among various risk groups, reinforcing the nomogram's substantial clinical value in predicting OS for PTL patients (P < 0.0001). SHAP value analysis clarified each variable's specific impact on the outcome. CONCLUSIONS: The nomogram provides a valuable instrument for clinicians to individualize OS predictions for PTL patients, addressing the unmet need for personalized prognostication in this rare malignancy.