作者
Liang Xia,Jiandong Mei,Ran Kang,Senyi Deng,Yaohui Chen,Yang Ying,Gang Feng,Yulan Deng,Fanyi Gan,Yi-Dan Lin,Qiang Pu,Lin Ma,Feng Lin,Yong Yuan,Yang Hu,Chenglin Guo,Hu Liao,Chengwu Liu,Yunke Zhu,Wen-Ping Wang,Zheng Liu,Yuyang Xu,Kaidi Li,Chuan Li,Qingyun Li,Ji He,Weizhi Chen,Xiaolong Zhang,Ying-Li Kou,Yun Wang,Zhu Wu,Guowei Che,Long-Qi Chen,Lunxu Liu
摘要
PURPOSE We assessed whether perioperative circulating tumor DNA (ctDNA) could be a biomarker for early detection of molecular residual disease (MRD) and prediction of postoperative relapse in resected non-small cell lung cancer (NSCLC). EXPERIMENTAL DESIGN Based on our prospective, multicenter cohort on dynamic monitoring of ctDNA in lung cancer surgery patients (LUNGCA), we enrolled 950 plasma samples obtained at three perioperative time points (before surgery, 3-day and 1-month after surgery) of 330 stage I-III NSCLC patients (LUNGCA-1), as a part of LUNGCA cohort. Using a customized 769-gene panel, somatic mutations in tumor tissues and plasma samples were identified with next-generation sequencing and utilized for ctDNA-based MRD analysis. RESULTS Preoperative ctDNA positivity was associated with lower recurrence-free survival (RFS) (HR=4.2; P<0.001). The presence of MRD (ctDNA positivity at postoperative 3-day and/or 1-month) was a strong predictor for disease relapse (HR=11.1; P<0.001). ctDNA-based MRD had a higher relative contribution to RFS prediction than all clinicopathological variables such as the TNM stage. Furthermore, MRD-positive patients who received adjuvant therapies had improved RFS over those not receiving adjuvant therapy (HR=0.3; P=0.008), whereas MRD-negative patients receiving adjuvant therapies had lower RFS than their counterparts without adjuvant therapy (HR=3.1; P<0.001). After adjusting for clinicopathological variables, whether receiving adjuvant therapies remained an independent factor for RFS in the MRD-positive population (P=0.002) but not in the MRD-negative population (P=0.283). CONCLUSIONS Perioperative ctDNA analysis is effective in early detection of MRD and relapse risk stratification of NSCLC, and hence could benefit NSCLC patient management.