Discover Oncology (Nov 2024)

Development and validation of survival nomograms for patients with anaplastic thyroid carcinoma: a SEER program-based study

  • Xinming Chen,
  • Pingwu Zhao,
  • Yunsheng He,
  • Kun Huang,
  • Pan Zhao,
  • Fengwan Liao,
  • Yang Liu

DOI
https://doi.org/10.1007/s12672-024-01537-5
Journal volume & issue
Vol. 15, no. 1
pp. 1 – 13

Abstract

Read online

Abstract Background We aimed to study the prognostic risk factors affecting patients with anaplastic thyroid carcinoma (ATC), develop a clinical prognostic model, and assess patient survival outcomes. Methods Patients with anaplastic thyroid carcinoma from 2000 to 2019 were selected from the Surveillance, Epidemiology, and End Results (SEER) Program to extract the clinical variables used for analysis. The dataset was divided into training (70%) and validation (30%) sets based on a 7:3 ratio. Univariate and LASSO regression analyses were performed on clinical variables from the training set to identify independent prognostic factors. Independent prognostic factors were determined by Univariate and lasso regression according to the clinical variables of the training set, and a nomogram model was established to construct a prognostic model based on the contribution degree of the predictors. The prognostic model was evaluated and internally verified by C-index, ROC curve and calibration curve. Results A total of 713 ATC patients were included in the SEER database. LASSO regression results indicated that age, marital status, race, tumor size, whether the primary lesion was limited to the thyroid gland, surgery, radiotherapy and chemotherapy, were associated with overall survival(OS) prognosis of ATC, and were used to construct nomograms. In the training cohort, the OS nomogram’s C-index was 0.708 (95% CI 0.672–0.745); in the internal validation cohort, the C-index was 0.677 (95% CI 0.620–0.735). ROC curves demonstrated that the OS nomogram exhibits excellent predictive accuracy and discriminative ability. Calibration curves indicated strong consistency between the OS nomogram’s predicted survival rates and actual survival rates. Conclusions We established a survival prediction model for ATC, which can assist clinicians in assessing patient prognosis and making personalized treatment decisions.

Keywords