Serum Extracellular Nano-Vesicles miR-153-3p to Identify Micronodular Lung Cancer from Sub-Centimeter Lung Nodules

Identification of malignancy from sub-centimeter lung nodules (LNs, <1 cm) is core for prevention and treatment of primary phases of lung cancer (LC). The study's purpose was to predict micro-nodular lung cancer (mnLC), such as adenocarcinoma in situ (AIS), micro-invasive adenocarcinoma (MIA), and invasive adenocarcinoma stage 1 (IA1) based on serum extracellular nano-vesicles (sEVs) miR-153-3p. sEVs-miR-153-3p was selected in this study and then used to investigate the expression and efficacy by RT-qPCR in the validation phase, followed by sEVs derived from patients of mnLC, benign lung nodule (BLN), and healthy people groups comprised of 135 recruiters. Further, the study established the prediction model which combined sEVs-miR-153-3p expression with multiple clinical-radiomics features by Logistic algorithms and was authenticated by the area under curve (AUC) phenomenon. Then, a simple-to-use nomogram was developed by Cox proportional-hazards regression modeling for predicting mnLC. Besides, in vitro analyses were performed to demonstrate miR-153-3p/ROCK1 axis in regulating biological mechanisms using LC cell lines. Results demonstrated that sEVs-miR-153-5p's expression values were higher in mnLC patients compared to BLN and healthy people. The prediction model for mnLC was successfully established, utilizing sEVs-miR-153-5p biomarker with significant clinical-radiomics features and yielding an AUC = 0.943 (95%CI: 0.898~0.989, P <0.0001). Moreover, the miR-153-3p and its targeted gene ROCK1 were confirmed as down-regulated in NSCLC cell lines and up-regulated expression respectively. Moreover, the miR-153-3p/ROCK1 axis took part in promoting the epithelial-mesenchymal transition (EMT) signaling pathway and regulated certain biological functions, such as proliferation, migration and invasion of LC cells. Therefore, the sEVs-miR-153-3p associated with radiomics-clinical features showed enormous potential to non-invasively identify malignancy from subcentimeter LNs. miR-153-3p could promote the genesis and progress of LC via EMT signaling pathway, which may serve as a therapeutic target.