实用医学杂志 ›› 2025, Vol. 41 ›› Issue (17): 2705-2714.doi: 10.3969/j.issn.1006-5725.2025.17.014

• 临床研究 • 上一篇    

非小细胞肺癌免疫治疗相关甲状腺功能障碍的危险因素分析及模型构建

曹玲春1,孟凡亮1(),盛晓安2   

  1. 1.安徽医科大学附属巢湖医院 呼吸内科 (安徽巢湖 238000 )
    2.安徽医科大学附属巢湖医院 肿瘤放疗科 (安徽巢湖 238000 )
  • 收稿日期:2025-04-29 出版日期:2025-09-10 发布日期:2025-09-05
  • 通讯作者: 孟凡亮 E-mail:13966337677@163.com
  • 基金资助:
    安徽省卫生健康科研项目(AHWJ2024BAd30010)

Risk factors and model construction of immune therapy⁃related thyroid dysfunction in non⁃small cell lung cancer

Lingchun CAO1,Fanliang MENG1(),Xiaoan SHENG2   

  1. Department of Respiratory Medicine,Chaohu Hospital Affiliated to Anhui Medical University,Chaohu 238000,Anhui,China
  • Received:2025-04-29 Online:2025-09-10 Published:2025-09-05
  • Contact: Fanliang MENG E-mail:13966337677@163.com

摘要:

目的 探讨非小细胞肺癌(NSCLC)患者免疫治疗后发生甲状腺功能障碍(irTD)的相关危险因素并构建预测模型。 方法 回顾性分析2019年1月至2024年6月在安徽医科大学附属巢湖医院收治的197例接受免疫治疗的NSCLC患者资料,按7∶3比例分为训练集(n = 137)和验证集(n = 60)。通过Lasso和logistic回归筛选危险因素并构建动态列线图模型,采用ROC曲线、校准曲线、决策曲线(DCA)及临床影响曲线(CIC)评估模型效能。 结果 多因素分析显示性别(OR = 0.172,95% CI:0.047 ~ 0.623)、M分期(OR = 2.919,95% CI:1.063 ~ 8.015)、ln(SII)(OR = 0.167,95% CI:0.066 ~ 0.423)、ALC(OR = 3.395,95% CI:1.493 ~ 7.716)、TSH(OR = 1.464,95% CI:1.126 ~ 1.904)是irTD发生的独立危险因素,据此构建的模型公式为logit(P) = 9.261 - 1.760 × 性别 + 1.071 × M分期 - 1.787 × ln(SII) + 1.222 × ALC + 0.381 × TSH,该模型在验证集中的AUC达0.832(95%CI:0.719 ~ 0.945),与训练集结果相近。校准曲线显示预测概率与实际观察值具有良好的一致性,DCA和CIC证实该模型具有较好的临床应用价值。 结论 该研究确定了NSCLC患者免疫治疗后发生irTD的关键危险因素,所建立的动态列线图模型(网页计算器:https://lingchun.shinyapps.io/dynnomapp/)能够有效识别irTD高风险人群,为临床决策提供可靠的工具。

关键词: 非小细胞肺癌, 免疫治疗, 甲状腺功能障碍, 列线图

Abstract:

Objective To explore the related risk factors for thyroid dysfunction (irTD) after immunotherapy in patients with non?small cell lung cancer (NSCLC) and to construct a predictive model. Methods A retrospective analysis was conducted on 197 NSCLC patients who received immunotherapy at Chaohu Hospital, Anhui Medical University between January 2019 and June 2024. The patients were divided into a training set (n = 137) and a validation set (n = 60) in a 7∶3 ratio. Risk factors were screened through Lasso and logistic regression, and a dynamic nomogram model was constructed. The model's performance was evaluated using ROC curve, calibration curve, decision curve analysis (DCA), and clinical impact curve (CIC). Results Multivariate analysis showed that Gender (OR = 0.172, 95% CI:0.047 ~ 0.623), M stage(OR = 2.919, 95% CI:1.063 ~ 8.015),ln(SII) (OR = 0.167, 95% CI:0.066 ~ 0.423), ALC (OR = 3.395, 95% CI:1.493 ~ 7.716), and TSH (OR = 1.464, 95% CI:1.126 ~ 1.904) were independent risk factors for the occurrence of irTD. The model formula based on these factors is: logit(P) = 9.261 - 1.760 × Gender + 1.071 × M stage - 1.787 × ln(SII) + 1.222 × ALC + 0.381 × TSH. The model achieved an AUC of 0.832(95%CI:0.719 ~ 0.945) in the validation set, which was similar to the results in the training set. The calibration curve demonstrated good consistency between the predicted probability and actual observed values. DCA and CIC confirmed that the model has good clinical applicability. Conclusion This study identifies key risk factors for the occurrence of irTD after immunotherapy in NSCLC patients. The dynamic nomogram model developed (web calculator: https://lingchun.shinyapps.io/dynnomapp/) can effectively identify high?risk populations for irTD, providing a reliable tool for clinical decision?making.

Key words: non?small cell lung cancer, immunotherapy, thyroid dysfunction, nomogram

中图分类号: