影像诊断与介入放射学

目的 周围型肺癌的影像组学特征与Ki-67表达水平之间的关系尚不清楚。本研究建立基于CT增强动脉期图像的影像组学标签预测周围型肺癌的Ki-67表达情况。方法 回顾性收集2017年5月—2020年11月行胸部CT增强扫描并在检查后2周内经病理证实、行Ki-67表达水平检测的117例周围型肺癌患者，其中男43例，女74例，年龄35～79岁（中位年龄54岁）。经手术病理证实，Ki-67高表达组54例，Ki-67低表达组63例，以7∶3的比例将患者分为训练组（n=82）和验证组（n=35）。使用ITK-SNAP于CT动脉期图像上手动勾画肺癌全肿瘤容积数据，A.K软件提取影像组学特征。采用LASSO回归模型进一步筛选特征并构建影像组学标签，并计算每例患者的影像组学评分，然后结合临床信息进行多因素Logistic回归分析，筛选出预测Ki-67水平的独立危险因素。在验证组和训练组中使用受试者工作特征曲线及曲线下面积（AUC）评价影像组学标签的预测性能。根据Hosmer-Lemeshow检验评估影像组学标签的校准度。采用决策曲线分析法（DCA）评估影像组学标签的临床价值。结果 从396个特征中选择7个影像组学特征，建立与Ki-67表达水平显著相关的影像组学标签。该模型在训练组中AUC为0.844(95%CI:0.725～0.964)，敏感度93%，特异度71%，校准度0.709。在验证组中，AUC为0.881(95%CI:0.756～0.954)，敏感度91%，特异度75%，校准度0.950。单因素Logistic回归分析显示Ki-67高表达与低表达两组间性别、年龄和吸烟差异均无统计学意义（P>0.05）。使用多因素Logistic回归模型，影像组学评分被认为是预测周围型肺癌Ki-67表达情况的独立预测因素。DCA显示阈值概率在0.03%～0.63%时，影像组学标签预测周围型肺癌Ki-67表达水平效能较优。结论 基于增强CT动脉期图像建立的影像组学标签有助于预测周围型肺癌病灶Ki-67的表达，无创评估肿瘤侵袭性和预后。

Objective To develop a radiomics signature based on arterial phase enhanced CT to estimate the Ki-67expression level in peripheral lung cancer. Methods A total of 117 patients(43 men, 74 women; age range: 35-79 years; median:54 years) with peripheral lung cancer underwent contrast-enhanced chest CT in our hospital from May 2016 to November 2019,were included. All of the peripheral lung cancers were pathologically confirmed and Ki-67 expression levels(63 low, 54 high) were assessed within 2 weeks after CT. The patients were divided into training(82) and validation(35) cohorts in a ratio of 7:3. ITK-SNAP was used to manually outline the total tumor volume of lung cancer on CT arterial phase images, and the radiomics features were extracted by A.K software. LASSO regression model was used to further screen features and construct radiomics labels. The radiomics score of each patient was calculated. Multi-factor logistic regression analysis was performed in combination with clinical information to screen out independent risk factors for predicting Ki-67 levels. The predictive accuracy of the radiomics signature was quantified by the area under receiver operator characteristic curve(AUC) in both the training and validation cohorts. The HosmerLemeshow test was performed to evaluate the calibration degree of the radiomics. We performed decision curve analysis(DCA) to assess the clinical usefulness of the radiomics signature. Results Seven radiomics features were chosen from 396 candidate features to build a radiomics label that significantly correlated with Ki-67 expression level. The model showed good calibration and discrimination in the training cohort, with an AUC of 0. 844( 95 % CI : 0. 725-0. 964), sensitivity of 93 %, specificity of 71 %, and calibration degree of 0.709. In the validation cohort, AUC was 0.881(95% CI: 0.756-0.954) with sensitivity of 91%, specificity of 75%, and calibration degree of 0.950. Univariate logistic regression analysis showed that there were no conspicuous differences in gender, age and smoking history between the high and low Ki-67 expressions( P >0.05). Using multivariate logistic regression model, radiomics signature was considered to be an independent predictor of Ki-67 expression level in peripheral lung cancer. DCA for the radiomics signature in the training cohort showed that if the threshold probability was between 0.03 and 0.63, then using the radiomics signature to predict Ki-67 expression situation added more benefit than assuming high or low Ki-67 expressions in all patients.Conclusion The radiomics signature based on arterial phase CT helps to noninvasively predict the expression of Ki-67 and thereby the invasiveness and prognosis of peripheral lung cancer.