Application of competing risk model in the prognostic prediction study of patients with follicular thyroid carcinoma.

BACKGROUND: Follicular thyroid carcinoma (FTC) is an indolent carcinoma. The cumulative incidence of death from patients with FTC and the nomogram built based on the competing risks model have not been described. METHODS: The data from patients diagnosed with primary FTC were identified and extracted from the surveillance, epidemiology, and end results (SEER) program (1988-2015). The cumulative incidence function was utilized to calculate the likelihood of death resulting from thyroid cancer and other causes, respectively. Gray's test was used to examine the difference in the cumulative incidence of death between the groups. A tenfold cross-validation was applied to assess the discrimination and calibration of the model. RESULTS: A total of 9210 patients diagnosed with primary FTC were included. The median follow-up time was 92 months (1-347 months). The 5-year, 10-year, and 20-year probabilities of death from FTC were 2.84%, 5.23%, and 8.61%, respectively. The age at diagnosis, sex, tumor size, pathological subtypes, tumor extension, lymph node involvement, as well as surgical and radiotherapy methods used, were related to the cumulative incidence of death. Multivariate analysis identified several risk factors for patient survival. The model behaved well in terms of performance. A nomogram based on the model allowed the prediction of the probability of death among patients with FTC. CONCLUSIONS: The prognosis of FTC is excellent. The likelihood of death caused by thyroid cancer increases with age. Male sex, tumors larger than 4 cm, invasion, extrathyroidal extension, lymph node involvement, and distant metastases increase the risk of dying of thyroid carcinoma. The nomogram constructed on the basis of the model is potentially useful for both clinicians and patients.