A dynamic nomogram and risk stratification system for predicting cancer-specific survival in patients with locally advanced differentiated thyroid cancer: a population-based study.

BACKGROUND

Locally advanced differentiated thyroid cancer (LADTC) refers to a severe stage of differentiated thyroid cancer (DTC) with a relatively poor prognosis. This study aimed to construct a dynamic nomogram and risk stratification system to predict cancer-specific survival (CSS) in patients with LADTC.

METHODS

A total of 4,856 patients diagnosed with LADTC from 2004 to 2020 were included from the Surveillance, Epidemiology, and End Results database. Least absolute shrinkage and selection operator (LASSO) and Cox regression analyses were utilized to identify variables and construct the dynamic nomogram. The performance of the nomogram was assessed using the concordance index (C-index), time-dependent receiver operating characteristic (ROC) curve, area under the curve (AUC), and calibration plot, while decision curve analysis (DCA) was conducted to evaluate clinical benefits. The improvement of the nomogram in comparison to the American Joint Committee on Cancer (AJCC) staging system was evaluated using the C-index, net reclassification index (NRI), and integrated discrimination improvement (IDI). A risk stratification system was established according to the total score of each patient in the nomogram.

RESULTS

Eight variables were identified to construct the nomogram. The C-index, time-dependent ROC curve, AUC, calibration plot, and DCA demonstrated the strong performance and clinical benefits of the nomogram. The C-index, NRI, and IDI indicated that the nomogram outperformed the AJCC staging system in prognostic prediction. The risk stratification system demonstrated the favorable ability to categorize patients with LADTC.

CONCLUSIONS

A dynamic nomogram and risk stratification system were constructed and validated to assist clinicians in evaluating prognostic risk and devising personalized treatment strategies for patients with LADTC.