Differentiating the Invasiveness of Lung Adenocarcinoma by Dual Energy CT Parameter

Although progression of lung adenocarcinoma (LUAD) depends on driver mutations, it is also affected by tumor microenvironment (TME), including vessels, immune cells and extracellular matrix (ECM). As major constituent of TME, ECM mediates interactions between cancer cells and stromal cells, promotes angiogenesis, epithelial-mesenchymal transition, causes metastasis and resistance to immune therapy. Along with the progression of LUAD histologic stages, from atypical adenomatous hyperplasia (AAH) and adenocarcinoma in situ (AIS) to minimally invasive adenocarcinoma (MIA), and finally to invasive adenocarcinoma (IA), the composition of the ECM changes a lot, which has some characterizations same as interstitial pulmonary fibrosis. Hence, identifying the pathological ECM status may help differentiating the invasiveness of lung adenocarcinomas. Based on the theory that in delay phase, the contrast medium is evenly distributed in the intravascular and extravascular-extracellular spaces and not entering the cell, extracellular volume (ECV) fraction is considered as a potential quantitative imaging parameter for ECM. It has been confirmed that ECV fraction is highly consistent with pathological fibrosis in cardiac and hepatic diseases. In other lesions with fibrosis such as pancreatic and thymic epithelial tumors, ECV fraction also has positive effects in malignancy prediction. It has been verified that ECV fraction is capable of differentiating lung cancer with benign lung lesions and classifying lung cancers into three subtypes. However, there has yet no study testified it as an invasion predictor. The core purpose of this study is to investigate whether ECV fraction can distinguish precancerous lesions from early-stage lung adenocarcinomas and compare it with other confirmed radiological features in prediction performance, which is clinically meaningful regarding optimal treatment selection and avoidance of unnecessary surgical procedures.