CT特征在预测非小细胞肺癌中甲状腺转录因子1表达状态的价值  

The Value of CT Features in Predicting the Expression Status of Thyroid Transcription Factor 1 in Non-Small Cell Lung Cancer

作　　者：闻敏 李红 刘松[1] 陈文萍 周正扬[1] WEN Min;LI Hong;LIU Song;CHEN Wen-ping;ZHOU Zheng-yang(Department of Radiology,Affiliated Drum tower Hospital,Medical School of Nanjing University,Nanjing 210008,Jiangsu Province,China)

机构地区：[1]南京大学医学院附属鼓楼医院影像医学科,江苏南京210008

出　　处：《中国CT和MRI杂志》2022年第6期42-45,共4页Chinese Journal of CT and MRI

摘　　要：目的探讨CT特征评估非小细胞肺癌甲状腺转录因子1表达状态的预测价值。方法回顾性分析2017年1月至2018年12月经手术病理证实并有甲状腺转录因子1免疫组化结果的146例非小细胞肺癌患者术前胸部CT图像,共评估了6个定量参数及17个定性参数,定量变量采用正态分布检验、偏态分布描述性统计、非参数方法Mann-Whitney u检验,定性变量采用χ^(2)检验或Fisher精确性检验。采用二元logistic回归构建预测TTF-1的多参数模型,通过ROC曲线比较各单因素及多参数模型预测值的诊断效能。采用组内相关系数(ICC)检测观察者间定量测量的一致性。结果定量参数中平均CT值、最小CT值、SD、长径、短径及定性参数中成分、支气管“充气”征、结节型/肿块型、胸膜增厚、阻塞性肺气肿有统计学意义(P<0.05)。ROC曲线分析示平均CT值、最小CT值、SD、长径及短径在预测TTF-1表达中诊断效能较好,其AUC值分别为0.682、0.696、0.728、0.764、0.745,多参数模型预测概率值AUC值为0.840,诊断效能最高。观察者间全部定量特征的测量一致性尚可(ICC0.40~0.75)或较高(ICC>0.75)。结论CT特征可以评估非小细胞肺癌TTF-1的表达状态,多参数模型诊断效能较单因素分析更高。Objective To explore the predictive value of CT features in evaluating the expression status of thyroid transcription factor 1 in non-small cell lung cancer.Methods A retrospective analysis of preoperative chest CT images of 146 patients with non-small cell lung cancer confirmed by surgery and pathology and had thyroid transcription factor 1(TTF-1)immunohistochemical results from January 2017 to December 2018.A total of 6 quantitative parameters and 17 qualitative parameters were evaluated.The quantitative variables were tested by normal distribution.Descriptive statistics of normal distribution test and Mann-Whitney u test were used for quatitative variables.χ^(2) test or Fisher's exact test were used for qualitative variables.Binary logistic regression was used to construct a multi-parameter model for predicting TTF-1,and the diagnostic efficacy of the predictive value of every single factor and multi-parameter model was compared through the ROC curve analysis.The intra-group correlation coefficient(ICC)was used to detect the consistency of quantitative measurements between observers.Results Among the quantitative parameters,the average CT value,minimum CT value,SD,long diameter,short diameter,and qualitative parameters of components,bronchial inflation sign,nodular/mass type,pleural thickening,and obstructive emphysema were statistically significant(all P<0.05).ROC curve analysis showed that the average CT value,minimum CT value,SD,long diameter and short diameter have good diagnostic performance in predicting the expression of TTF-1,and the AUC values are 0.682,0.696,0.728,0.764,and 0.745,respectively.For multi-parameter model prediction,the AUC value was 0.840,and the diagnostic efficiency is the highest.The measurement consistency of all quantitative characteristics between observers was fair(ICC0.40-0.75)or higher(ICC>0.75).Conclusion CT features can evaluate the expression status of TTF-1 in non-small cell lung cancer,and the diagnostic efficiency of the multi-parameter model was higher than that of sing

关 键 词：非小细胞肺癌 甲状腺转录因子1 体层摄影术 X线计算机 

分 类 号：R734.2[医药卫生—肿瘤]