Perioperative ctDNA dynamics predict outcomes in resectable NSCLC using an ultrasensitive and cost-effective assay

Recurrence after curative surgery is a major challenge in early-stage non-small cell lung cancer (NSCLC), as conventional imaging often fails to detect minimal residual disease (MRD). We prospectively analyzed 119 resected NSCLC to investigate the clinical utility of circulating tumor DNA (ctDNA) for predicting recurrence by PErsonalized Analysis of Cancer (PEAC) system. ctDNA dynamics were correlated with disease-free survival (DFS), overall survival (OS), and recurrence. Intraoperative blood sampling yielded the highest concordance with tumor tissue mutations. Baseline ctDNA negativity predicted better DFS (HR = 0.18). Longitudinal monitoring detected MRD in 48.1% of relapsed patients and anticipated radiographic recurrence by a median of 6.53 months. Persistently negative ctDNA correlated with favorable outcomes, while dynamic conversion to positivity was associated with recurrence. A recurrence prediction model combining baseline ctDNA and nodal stage achieved robust discrimination for 1-, 2-, and 3-year DFS (AUC = 0.73, 0.76, 0.72). PEAC-based dynamic ctDNA monitoring enables early detection of postoperative recurrence in NSCLC and serves as a valuable tool for risk stratification in adjuvant treatment decision-making. Integration of ctDNA with clinical parameters improves recurrence prediction and may guide personalized postoperative management. (EVOLUTION, GASTO1058; ClinicalTrials.gov identifier NCT04238130; registered on 18 December 2019)