Purpose To develop a nomogram combined radiomics,clinical and CT morphographic features for the prediction of the acquired T790M mutation in patients with advanced non-small cell lung cancer with resistance after the duration of first-line epidermal growth factor receptor(EGFR)-tyrosine kinase inhibitor(TKI)treatment.Materials and Methods The chest CT images and clinical data of 116 patients who underwent secondary biopsy after disease progression during first-line TKI treatment for advanced non-small cell lung cancer with EGFR sensitive mutations,from December 2016 to December 2022 in Shaoxing People's Hospital were retrospectively analyzed.All patients were randomly divided into a training cohort(n=81)and a validation cohort(n=35),at a ratio of 7︰3.The regions of interest of the lesion were delineated,and radiomics features were extracted.Feature selection was performed via the maximum relevance minimum redundancy and the least absolute shrinkage and selection operator methods.Clinical features were selected via univariate and multivariate Logistic regression.A nomogram combined clinical and radiomics features was subsequently constructed.The predictive ability of the combined model was evaluated via receiver operator characteristic curve,calibration curves and decision curve analysis.Results The progression-free survival of first-line EGFR-TKIs(OR=1.086,P=0.041),initial EGFR profile(OR=0.280,P=0.021),vascular convergence(OR=4.050,P=0.036)and air bronchogram(OR=3.265,P=0.030)were highly correlated with the acquired T790M mutation.The combined model demonstrated good predictive performance for acquired T790M mutation,both in the training set(AUC=0.867,95%CI 0.790-0.944)and the validation set(AUC=0.895,95%CI 0.786-1.000).The calibration curve showed good calibration power,and the decision curve analysis demonstrated a significant net benefit.Conclusion A radiomics-clinical nomogram based on CT radiomics has the potential to predict acquired T790M mutation and could be a complementary tool for T790M mutation detection after resistance to first-or second-generation EGFR-TKIs.

Purpose To develop a nomogram combined radiomics,clinical and CT morphographic features for the prediction of the acquired T790M mutation in patients with advanced non-small cell lung cancer with resistance after the duration of first-line epidermal growth factor receptor(EGFR)-tyrosine kinase inhibitor(TKI)treatment.Materials and Methods The chest CT images and clinical data of 116 patients who underwent secondary biopsy after disease progression during first-line TKI treatment for advanced non-small cell lung cancer with EGFR sensitive mutations,from December 2016 to December 2022 in Shaoxing People's Hospital were retrospectively analyzed.All patients were randomly divided into a training cohort(n=81)and a validation cohort(n=35),at a ratio of 7︰3.The regions of interest of the lesion were delineated,and radiomics features were extracted.Feature selection was performed via the maximum relevance minimum redundancy and the least absolute shrinkage and selection operator methods.Clinical features were selected via univariate and multivariate Logistic regression.A nomogram combined clinical and radiomics features was subsequently constructed.The predictive ability of the combined model was evaluated via receiver operator characteristic curve,calibration curves and decision curve analysis.Results The progression-free survival of first-line EGFR-TKIs(OR=1.086,P=0.041),initial EGFR profile(OR=0.280,P=0.021),vascular convergence(OR=4.050,P=0.036)and air bronchogram(OR=3.265,P=0.030)were highly correlated with the acquired T790M mutation.The combined model demonstrated good predictive performance for acquired T790M mutation,both in the training set(AUC=0.867,95%CI 0.790-0.944)and the validation set(AUC=0.895,95%CI 0.786-1.000).The calibration curve showed good calibration power,and the decision curve analysis demonstrated a significant net benefit.Conclusion A radiomics-clinical nomogram based on CT radiomics has the potential to predict acquired T790M mutation and could be a complementary tool for T790M mutation detection after resistance to first-or second-generation EGFR-TKIs.