Fei Zhou, Chunxia Su, Wenhua Liang, Li-Di Xu, Wei Tao, Juanjuan Qian, Shidong Xu, Dan Liu, Rui Fan, Yaxi Zhang, Yiqian Liu, Lin Wu, Xiangning Fu, Xiaodong Zhang, Ning Shen, Di Ge, Qun Chen, Weimin Mao, Lin Xu, Chun Chen, Bing Hu, Guoguang Shao, Jian Hu, Jian Zhao, Xiaoqing Liu, Zhidong Liu, Zheng Wang, Zemin Xiao, Taiqian Gong, Wen Lin, Xingya Li, Feng Ye, Yang Liu, Haitao Ma, Yunchao Huang, Jianying Zhou, Zhonglin Wang, Junke Fu, Lieming Ding, Li Mao, Weizhi Chen, Jianxing He, Caicun Zhou
Molecular residual disease (MRD), detected through circulating tumor DNA, has emerged as a promising biomarker for surveillance in patients with early-stage non-small cell lung cancer (NSCLC) following curative-intent resection. Here we report the MRD analysis from the phase 3 EVIDENCE trial, which includes patients with stage II-IIIA resected EGFR-mutated NSCLC who have received either adjuvant icotinib or chemotherapy. A total of 1,352 plasma samples from 175 patients are analyzed using the MinerVa Prime assay, a personalized tumor-informed MRD platform. At the landmark timepoint, MinerVa Prime detects MRD positivity in 35.8% (38/106) of patients with stage III disease and 14.7% (10/68) of patients with stage II disease, with a longitudinal positivity rate of 47.4%. MRD-positive status is significantly correlated with inferior disease-free survival (DFS), both at landmark (hazard ratio [HR]: 4.44, P < 0.001) and during longitudinal monitoring (HR: 7.82, P < 0.001). Icotinib confers DFS benefits over chemotherapy in both MRD subgroups, enhancing MRD clearance in MRD-positive patients while reducing molecular recurrence in landmark MRD-negative patients. Longitudinal MRD shows a 91.3% negative predictive value, with a median lead time of 169 days prior to clinical recurrence. Among serial monitoring timepoints, MRD status at 24 weeks post-randomization demonstrates the highest prognostic value.