Abstract
BACKGROUND: Preoperative tumor invasiveness in clinical stage T(1)N(0)M(0) lung adenocarcinoma is critical for optimal surgical procedure. The aim of the present study was to evaluate the relationship between the ground-glass opacity component (GGOc) / solid component (Sc) proportion measured using three-dimensional (3D) computer-quantified computer tomography (CT) number analysis to explore radiographic features for invasiveness prediction in cT(1)N(0)M(0) lung adenocarcinomas. METHODS: A total of 375 surgically resected cT(1)N(0)M(0) lung adenocarcinoma patients were included. The relativity between the GGOc/Sc proportion and lepidic growth pattern percentage was assessed using Spearman's rank analysis. Multiple logistic regression analysis was used to determine independent factors from radiographic features for tumor invasiveness. Prediction probability for tumor invasiveness was analysed using a receiver operating characteristic curve (ROC). RESULTS: We found that the GGOc proportion was positively correlated with lepidic growth pattern percentage (r = 0.67, P < 0.01), while the Sc proportion was negatively correlated with it (r = - 0.74, P < 0.01). Multivariate analysis showed that tumor size and Sc proportion were identified as independent predictors for tumor invasiveness. The area under the ROC curve (AUC) of Sc proportion was 0.875, which was higher than that of tumor size (0.750) (P < 0.001), and had no significant difference with that of combination of these two factors (0.884) (P = 0.28). CONCLUSIONS: The GGOc/Sc proportion measured using 3D computer-quantified CT number analysis reflects the lepidic growth pattern percentage in tumors, and the Sc proportion may be an important factor for the prediction of tumor invasiveness in cT(1)N(0)M(0) lung adenocarcinoma.