列线图
医学
接收机工作特性
逻辑回归
肿瘤科
内科学
逐步回归
鼻咽癌
曲线下面积
Lasso(编程语言)
不利影响
放射治疗
计算机科学
万维网
作者
Mengyuan Liu,Zheran Liu,Shuangshuang He,Yiyan Pei,Shihong Xu,Junyou Ge,Yan Qing,Youneng Wei,Ye Chen,Ping Ai,Xingchen Peng
标识
DOI:10.3389/fonc.2025.1539514
摘要
Objective: To predict the incidence of immune-related Adverse Events (irAEs) in patients with recurrent or metastatic Nasopharyngeal Carcinoma (NPC) treated with Programmed Death-Ligand 1 (PD-L1) inhibitors, this study developed and validated nomogram models incorporating demographic, clinical, and biological variables. Methods: Data from 153 NPC patients were analyzed, incorporating variables including age, sex, Body Mass Index (BMI), clinical stage, and biomarkers. Predictive models were constructed using multivariable logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO) regression, and Ridge regression. The models' performance was evaluated using Receiver Operating Characteristic (ROC) curves, calibration curves, and Decision Curve Analysis (DCA). Internal validation was conducted through k-fold cross-validation. Results: Independent predictors of irAEs included PD-L1, Free Thyroxine (FT4), Sodium (Na), and lymphocyte counts. Of the three models, the stepwise regression model performed best, with an area under the curve (AUC) of 0.78. Calibration curves showed a strong correlation between predicted and observed outcomes, and DCA demonstrated high clinical utility. Conclusion: The nomogram models effectively predict irAEs in NPC patients treated with PD-L1 inhibitors. Early identification of patients with elevated PD-L1, abnormal FT4, Na, or irregular lymphocyte counts allows for closer monitoring and personalized treatment, potentially improving outcomes. Further research is required to confirm these findings across other cancer types and therapies.
科研通智能强力驱动
Strongly Powered by AbleSci AI