Lung cancer remains the leading cause of cancer-related mortality worldwide, with recurrence or cancer-related death occurring even after curative-intent therapy. In this multicenter case–control study, we developed and validated a noninvasive urinary extracellular vesicle–derived microRNA (miRNA)-based assay for early-stage lung cancer detection and explored its utility for prognosis prediction and recurrence monitoring. Urine samples from 278 patients with lung cancer (half with early-stage disease) and 213 non-cancer controls were analyzed using small RNA sequencing and machine learning. The detection model showed high performance, with AUCs of 0.942 in the training and 0.941 in the test set. Sensitivity and specificity for early-stage lung cancer were 82.0% and 92.5% in the training set, and 88.2% and 87.0% in the test set, respectively. Twelve miRNAs decreased after surgery and rose again at recurrence. Eleven miRNAs were significantly associated with recurrence-free survival, and a 3-miRNA prognostic panel stratified patients into high- and low-risk groups with significantly different outcomes. Although systemic differences in age and sample storage between cohorts were observed, these findings suggest that urinary EV–miRNA profiling may be useful for early detection of lung cancer in a case–control setting and underscore its potential relevance for prognosis prediction and recurrence monitoring.
Kiritani et al. (Sat,) studied this question.