Objective: 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. Method: 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 ¹⁸F-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. Results: One hundred and nineteen patients with lung adenocarcinoma were included in our study. The whole group was divided into three datasets: a training set (n = 96), a validation set (n = 11), and a testing set (n = 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: 0.619~0.843), 0.750 (95% CI: 0.248~1.000), 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 (p = 0.377 and 0.861, respectively). Conclusions: Our model combining ¹⁸F-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.

Purpose To study the relationship between standardized 18F-FDG PET/CT radiomic features and clinicopathological variables and programmed death ligand-1 (PD-L1) expression status in non-small cell lung cancer (NSCLC) patients. Methods 58 NSCLC patients with preoperative 18F-FDG PET/CT scans and postoperative results of PD-L1 expression were retrospectively analysed. A standardized, open-source software was used to extract 86 radiomic features from PET and low-dose CT images. Univariate analysis and multivariate logistic regression were used to find independent predictors of PD-L1 expression. The Area Under the Curve (AUC) of receiver operating characteristic (ROC) curve was used to compare the ability of variables and their combination in predicting PD-L1 expression. Results Multivariate logistic regression resulted in the PET radiomic feature GLRLM_LGRE (Odds Rate (OR): 0.300 vs 0.114, 95% confidence interval (CI): 0.096–0.931 vs 0.021–0.616, in NSCLC and adenocarcinoma respectively) and the CT radiomic feature GLZLM_SZE (OR: 3.338 vs 7.504, 95%CI: 1.074–10.375 vs 1.382–40.755, in NSCLC and adenocarcinoma respectively), being independent predictors of PD-L1 status. In NSCLC group, after adjusting for gender and histology, the PET radiomic feature GLRLM_LGRE (OR: 0.282, 95%CI: 0.085–0.936) remained an independent predictor for PD-L1 status. In the adenocarcinoma group, when adjusting for gender the PET radiomic feature GLRLM_LGRE (OR: 0.115, 95%CI: 0.021–0.631) and the CT radiomic feature GLZLM_SZE (OR: 7.343, 95%CI: 1.285–41.965) remained associated with PD-L1 expression. Conclusion NSCLC and adenocarcinoma with PD-L1 expression show higher tumour heterogeneity. Heterogeneity-related 18F-FDG PET and CT radiomic features showed good ability to non-invasively predict PD-L1 expression.