Page 16 - 南京医科大学自然版

南京医科大学学报（自然科学版） 第46卷第8期
·1122 · Journal of Nanjing Medical University（Natural Sciences） 2026年8月

·专题研究：肿瘤·

基于CT影像组学评估PD⁃1免疫联合治疗非小细胞肺癌疗效的

初步研究

张林岑 ，王 芬 ，袁 梅 1*
3
1，2
南京医科大学第一附属医院放射科，江苏 南京 210029；南京医科大学附属逸夫医院放射科，江苏 南京 211100；
1 2
南京医科大学附属淮安第一医院影像中心，江苏 淮安 223300
3

［摘 要］ 目的：基于实体瘤免疫治疗疗效评价标准（immune response evaluation criteria in solid tumors，iRECIST）构建影像组
学模型，以预测非小细胞肺癌（non⁃small cell lung cancer，NSCLC）患者对免疫联合治疗的响应情况，从而在治疗前或早期阶段
实现临床获益人群的精准识别。方法：纳入75例经病理证实为NSCLC并接受免疫联合治疗的患者，所有患者均在治疗前1个
月内接受胸部CT检查，基于iRECIST对患者免疫联合治疗4~6个疗程（12~18周）后的疗效进行评估，将患者分为治疗敏感组
和非敏感组。患者按 2∶1 的比例随机分为训练集（n=50）和测试集（n=25），分别建立临床特征和影像组学模型。采用受试者
工作特征（receiver operating characteristic，ROC）曲线、曲线下面积（area under curve，AUC）、准确率（accuracy，ACC）及决策曲线
分析（decision curve analysis，DCA）对各模型的预测性能和临床应用价值进行比较。结果：经过特征提取与筛选，共得到5个组
学特征，包括2个一阶特征及3个纹理特征，经过特征融合和筛选，同时为了降低过拟合，进行3折交叉验证，最终朴素贝叶斯
模型最稳定，训练集及测试集AUC分别为0.675、0.647，ACC分别为0.720、0.720。结论：影像组学预测NSCLC患者对免疫联合
治疗的疗效有一定的潜力和临床应用价值。
［关键词］ 影像组学；非小细胞肺癌；PD⁃1；免疫联合治疗
［中图分类号］ R734.2；R814.42 ［文献标志码］ A ［文章编号］ 1007⁃4368（2026）08⁃1122⁃10
doi：10.7655/NYDXBNSN260228

Preliminary evaluation of PD⁃1 immunotherapy combination therapy efficacy in non⁃small
cell lung cancer using CT⁃based radiomics

1，2
3
ZHANG Lincen ，WANG Fen ，YUAN Mei 1*
2
1 Department of Radiology，the First Affiliated Hospital of Nanjing Medical University，Nanjing 210029；Department
of Radiology，Sir Run Run Hospital of Nanjing Medical University，Nanjing 211100；Department of Medical
3
Imaging，Huai’an First People’s Hospital Affiliated to Nanjing Medical University，Huai’an 223300，China

［Abstract］ Objective：To develop a radiomics ⁃ based model based on the immune response evaluation criteria in solid tumors
（iRECIST）for predicting the response of patients with non ⁃ small cell lung cancer（NSCLC）to combined immunotherapy，so as to
precisely screen patients who may derive clinical benefit at the pre⁃treatment or early treatment stage. Methods：A total of 75 patients
with pathologically confirmed NSCLC and received combined immunotherapy were enrolled. All patients underwent chest computed
tomography（CT）within one month before treatment. Based on the iRECIST，treatment efficacy after 4 ⁃ 6 cycles（12 ⁃ 18 weeks）of
immunotherapy combination therapy was assessed，and patients were classified into the treatment⁃sensitive and non⁃sensitive groups.
The patients were randomly divided into a training cohort（n=50）and a testing cohort（n=25）at a ratio of 2∶1. Clinical feature models
and radiomics prediction models were developed separately. The predictive performance and clinical utility of the models were
compared using receiver operating characteristic（ROC）curves，the area under the curve（AUC），accuracy（ACC），and decision curve
analysis（DCA）. Results：After feature extraction and selection，a total of five radiomics features were obtained，including two first⁃
order features and three texture features. After feature fusion and screening，and in order to reduce over⁃fitting，through three⁃fold cross⁃

［基金项目］ 医学影像人工智能专项研究基金；南京医学会第十四届放射学分会（序号13）
通信作者（Corresponding author），E⁃mail：yuanmeijiangsu@163.com（ORCID：0000⁃0002⁃5287⁃9198）
∗