Zhang Ruiyun, Shi Kuangyu, Hohenforst-Schmidt Wolfgang, Steppert Claus, Sziklavari Zsolt, Schmidkonz Christian, Atzinger Armin, Hartmann Arndt, Vieth Michael, Förster Stefan
Institute of Pathology, Medizincampus Oberfranken, Klinikum Bayreuth, Friedrich-Alexander-Universität Erlangen-Nürnberg, 95445 Bayreuth, Germany.
Institute of Pathology, Universitätsklinikum Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg, 91054 Erlangen, Germany.
Cancers (Basel). 2023 Jul 19;15(14):3684. doi: 10.3390/cancers15143684.
Considering the essential role of KRAS mutation in NSCLC and the limited experience of PET radiomic features in KRAS mutation, a prediction model was built in our current analysis. Our model aims to evaluate the status of KRAS mutants in lung adenocarcinoma by combining PET radiomics and machine learning.
Patients were retrospectively selected from our database and screened from the NSCLC radiogenomic dataset from TCIA. The dataset was randomly divided into three subgroups. Two open-source software programs, 3D Slicer and Python, were used to segment lung tumours and extract radiomic features from F-FDG-PET images. Feature selection was performed by the Mann-Whitney U test, Spearman's rank correlation coefficient, and RFE. Logistic regression was used to build the prediction models. AUCs from ROCs were used to compare the predictive abilities of the models. Calibration plots were obtained to examine the agreements of observed and predictive values in the validation and testing groups. DCA curves were performed to check the clinical impact of the best model. Finally, a nomogram was obtained to present the selected model.
One hundred and nineteen patients with lung adenocarcinoma were included in our study. The whole group was divided into three datasets: a training set ( = 96), a validation set ( = 11), and a testing set ( = 12). In total, 1781 radiomic features were extracted from PET images. One hundred sixty-three predictive models were established according to each original feature group and their combinations. After model comparison and selection, one model, including wHLH_fo_IR, wHLH_glrlm_SRHGLE, wHLH_glszm_SAHGLE, and smoking habits, was validated with the highest predictive value. The model obtained AUCs of 0.731 (95% CI: 0.6190.843), 0.750 (95% CI: 0.2481.000), and 0.750 (95% CI: 0.448~1.000) in the training set, the validation set and the testing set, respectively. Results from calibration plots in validation and testing groups indicated that there was no departure between observed and predictive values in the two datasets ( = 0.377 and 0.861, respectively).
Our model combining F-FDG-PET radiomics and machine learning indicated a good predictive ability of KRAS status in lung adenocarcinoma. It may be a helpful non-invasive method to screen the KRAS mutation status of heterogenous lung adenocarcinoma before selected biopsy sampling.
鉴于KRAS突变在非小细胞肺癌中的重要作用以及PET影像组学特征在KRAS突变方面的经验有限，我们在当前分析中构建了一个预测模型。我们的模型旨在通过结合PET影像组学和机器学习来评估肺腺癌中KRAS突变体的状态。
从我们的数据库中回顾性选择患者，并从TCIA的NSCLC放射基因组数据集中进行筛选。该数据集被随机分为三个亚组。使用两个开源软件程序3D Slicer和Python对肺肿瘤进行分割，并从F-FDG-PET图像中提取影像组学特征。通过曼-惠特尼U检验、斯皮尔曼等级相关系数和RFE进行特征选择。使用逻辑回归构建预测模型。使用ROC曲线下面积（AUC）来比较模型的预测能力。获得校准图以检查验证组和测试组中观察值与预测值的一致性。进行决策曲线分析（DCA）以检查最佳模型的临床影响。最后，获得列线图以展示所选模型。
我们的研究纳入了119例肺腺癌患者。整个组被分为三个数据集：训练集（n = 96）、验证集（n = 11）和测试集（n = 12）。总共从PET图像中提取了1781个影像组学特征。根据每个原始特征组及其组合建立了163个预测模型。经过模型比较和选择，一个包括wHLH_fo_IR、wHLH_glrlm_SRHGLE、wHLH_glszm_SAHGLE和吸烟习惯的模型被验证具有最高的预测价值。该模型在训练集、验证集和测试集中的AUC分别为0.731（95%CI：0.6190.843）、0.750（95%CI：0.2481.000）和0.750（95%CI：0.448~1.000）。验证组和测试组校准图的结果表明，两个数据集中观察值与预测值之间没有偏差（分别为P = 0.377和0.861）。
我们结合F-FDG-PET影像组学和机器学习的模型显示出对肺腺癌中KRAS状态具有良好的预测能力。它可能是在进行活检采样前筛选异质性肺腺癌KRAS突变状态的一种有用的非侵入性方法。
