B Sacconi1, M Anzidei2, A Leonardi2, F Boni2, L Saba3, R Scipione2, M Anile4, M Rengo5, F Longo6, M Bezzi2, F Venuta4, A Napoli2, A Laghi5, C Catalano2. 1. Department of Radiological, Oncological and Anatomopathological Sciences - Radiology, "Sapienza" University of Rome, Rome, Italy. Electronic address: beatrice.sacconi@fastwebnet.it. 2. Department of Radiological, Oncological and Anatomopathological Sciences - Radiology, "Sapienza" University of Rome, Rome, Italy. 3. Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), di Cagliari - Polo di Monserrato, Monserrato, Italy. 4. Department of Thoracic Surgery, "Sapienza" University of Rome, Rome, Italy. 5. Department of Radiological, Oncological and Anatomopathological Sciences - Radiology, "Sapienza" University of Rome Polo Pontino, Latina, Italy. 6. Department of Radiological, Oncological and Anatomopathological Sciences - Oncology, "Sapienza" University of Rome, Rome, Italy.
Abstract
AIM: To investigate the correlation between conventional computed tomography (CT) features, quantitative texture analysis (QTA), epidermal growth factor receptor (EGFR) mutations, and survival rates in patients with lung adenocarcinoma. MATERIALS AND METHODS: Sixty-eight patients were evaluated for conventional CT features and QTA in this retrospective study. A multiple logistic regression analysis and receiver operating characteristics (ROC) curve analysis versus death and EGFR status was performed for CT features and QTA in order to assess correlation between CT features, QTA, EGFR mutations, and survival rates. A p-value <0.05 was regarded to indicate a statistically significant association. RESULTS: An EGFR mutation was identified in 26/68 tumours (38.2%). A negative association was found between EGFR mutation and emphysema (p < 0.0001) whereas a positive correlation was found with necrosis (p=0.017), air bronchogram (p=0.0304), and locoregional infiltration (p=0.0018). Mean, standard deviation, and skewness were found to have significant correlation with EGFR mutation (p=0.0001; p=0.0001; p=0.0459; Fig 3). The only parameter correlated with the event death was entropy (r=0.2708; p=0.0329). CONCLUSION: Both qualitative and quantitative analysis disclosed potential associations between CT features and QTA parameters, EGFR mutations and prognosis; these correlations need to be confirmed in larger studies to be used as imaging biomarkers in the management of patients affected by lung adenocarcinoma.
AIM: To investigate the correlation between conventional computed tomography (CT) features, quantitative texture analysis (QTA), epidermal growth factor receptor (EGFR) mutations, and survival rates in patients with lung adenocarcinoma. MATERIALS AND METHODS: Sixty-eight patients were evaluated for conventional CT features and QTA in this retrospective study. A multiple logistic regression analysis and receiver operating characteristics (ROC) curve analysis versus death and EGFR status was performed for CT features and QTA in order to assess correlation between CT features, QTA, EGFR mutations, and survival rates. A p-value <0.05 was regarded to indicate a statistically significant association. RESULTS: An EGFR mutation was identified in 26/68 tumours (38.2%). A negative association was found between EGFR mutation and emphysema (p < 0.0001) whereas a positive correlation was found with necrosis (p=0.017), air bronchogram (p=0.0304), and locoregional infiltration (p=0.0018). Mean, standard deviation, and skewness were found to have significant correlation with EGFR mutation (p=0.0001; p=0.0001; p=0.0459; Fig 3). The only parameter correlated with the event death was entropy (r=0.2708; p=0.0329). CONCLUSION: Both qualitative and quantitative analysis disclosed potential associations between CT features and QTA parameters, EGFR mutations and prognosis; these correlations need to be confirmed in larger studies to be used as imaging biomarkers in the management of patients affected by lung adenocarcinoma.
Authors: Hannah Able; Amber Wolf-Ringwall; Aaron Rendahl; Christopher P Ober; Davis M Seelig; Chris T Wilke; Jessica Lawrence Journal: PLoS One Date: 2021-08-17 Impact factor: 3.240