Scholarlink

Background <p>Patients with metastatic thyroid cancer (TC) exhibit marked heterogeneity in clinical outcomes, yet reliable tools for risk stratification and treatment guidance remain unavailable in routine practice.</p> Methods <p>Patients diagnosed with metastatic TC were identified from the SEER database. Prognostic clinicopathological variables were screened using Cox proportional hazards analysis and integrated into a random survival forest (RSF) model for survival prediction and risk stratification. Model performance was evaluated using the concordance index (C-index) and time-dependent receiver operating characteristic curves. External validation was conducted using an independent institutional cohort.</p> Results <p>Surgery of the primary tumor, histological subtype, and radiotherapy were identified as the most influential determinants of overall survival (OS). The RSF model demonstrated superior discriminative performance compared with the conventional Cox model, achieving C-index value of 0.837 (95%CI: 0.805–0.854) in the internal test cohort. The internal test cohort showed 1-, 3-, and 5-year AUC values of 0.913, 0.943, and 0.916, respectively, whereas the external validation cohort showed corresponding AUC values of 0.912, 0.891, and 0.873, respectively. RSF-based risk stratification effectively separated patients into low- and high-risk groups with significantly different OS outcomes (<i>P</i> &lt; 0.001). Notably, although high-risk patients were characterized by advanced disease features and inferior baseline prognosis, subgroup analyses revealed that active treatment interventions, particularly primary tumor surgery, were associated with improved overall survival within this high-risk population.</p> Conclusions <p>This RSF-based prognostic model enables accurate risk stratification in metastatic TC and uncovers clinically actionable heterogeneity among high-risk patients. By identifying high-risk individuals with treatment-associated survival differences, the model provides a valuable framework for individualized therapeutic decision-making.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Risk stratification and treatment-associated survival analysis in distant metastatic thyroid cancer using a random survival forest model

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月30日

查看全文
  • Hong Luo,
  • Yun Zhao

摘要

Background

Patients with metastatic thyroid cancer (TC) exhibit marked heterogeneity in clinical outcomes, yet reliable tools for risk stratification and treatment guidance remain unavailable in routine practice.

Methods

Patients diagnosed with metastatic TC were identified from the SEER database. Prognostic clinicopathological variables were screened using Cox proportional hazards analysis and integrated into a random survival forest (RSF) model for survival prediction and risk stratification. Model performance was evaluated using the concordance index (C-index) and time-dependent receiver operating characteristic curves. External validation was conducted using an independent institutional cohort.

Results

Surgery of the primary tumor, histological subtype, and radiotherapy were identified as the most influential determinants of overall survival (OS). The RSF model demonstrated superior discriminative performance compared with the conventional Cox model, achieving C-index value of 0.837 (95%CI: 0.805–0.854) in the internal test cohort. The internal test cohort showed 1-, 3-, and 5-year AUC values of 0.913, 0.943, and 0.916, respectively, whereas the external validation cohort showed corresponding AUC values of 0.912, 0.891, and 0.873, respectively. RSF-based risk stratification effectively separated patients into low- and high-risk groups with significantly different OS outcomes (P < 0.001). Notably, although high-risk patients were characterized by advanced disease features and inferior baseline prognosis, subgroup analyses revealed that active treatment interventions, particularly primary tumor surgery, were associated with improved overall survival within this high-risk population.

Conclusions

This RSF-based prognostic model enables accurate risk stratification in metastatic TC and uncovers clinically actionable heterogeneity among high-risk patients. By identifying high-risk individuals with treatment-associated survival differences, the model provides a valuable framework for individualized therapeutic decision-making.