首都医科大学

Article https://doi.org/10.1038/s41467-026-72983-8 Multi-phase hybrid metabolomics framework identifies clinically applicable plasma signatures for early detection of gastric cancer Liyi Bai 1,2,3,4,10, Fayong Hu5,10, Weiqin Zhang 6, Huanqin Peng6, Haowen Peng6, Huan Li6, Xu Zhu7 , Yibin Xie 8 , Shutian Zhang 1,2,3,4 & Li Min 1,2,3,4,9 Plasma metabolomics offers significant potential for non-invasive biomarker discovery in gastric cancer (GC), yet conventional analytical workflows face challenges in absolute quantification and biological interpretability, hindering clinical translation. Here we present an innovative multi-phase hybrid frame- work integrating untargeted metabolomics with relative- and absolute- quantitative targeted metabolomics, coupled with a custom interpretability- driven algorithm for de novo biomarker identification. We performmetabolic profiling on 1,706 plasma samples from multicenter cohorts, identifying 84 keymetabolites significantly enriched in caffeinemetabolism and primary bile acid biosynthesis during the relative quantitation phase. By applying the cus- tom algorithm to absolute quantitation data, we establish a 12-metabolite panel covering multiple functional metabolic modules. Machine learning- based diagnostic models using this signature achieve an area under the curve of 0.951 in validation cohort. Together, our study provides a robust and interpretable framework for translational metabolomics and establishes a GC detection biomarker panel, laying the foundation for future mechanistic research and clinical application. Gastric cancer (GC) ranks fifth in global incidence among common malignancies and stands as the third leading cause of cancer-related death worldwide1. The five-year survival rate for patients with early gastric cancer (EGC) exceeds 90%2, but falls below 10% in those with advanced gastric cancer (AGC)3. This striking survival disparity underscores the imperative for early GC detection. While endoscopic examination has been frequently used for early detection of GC due to its diagnostic reliability4, the inherent procedural invasiveness and requirement for specialized operator skills constrain its broad clinical implementation. Conventional serum tumor markers (TMs) represent Received: 22 September 2025 Accepted: 24 April 2026 Check for updates 1Department of Gastroenterology, Beijing Friendship Hospital, Capital Medical University, Beijing, China. 2State Key Laboratory of Digestive Health, Beijing, China. 3National Clinical ResearchCenter for Digestive Diseases, Beijing, China. 4Beijing Key Laboratory of Early Gastrointestinal CancerMedicine and Medical Devices, Beijing, China. 5Department of Gastrointestinal Surgery, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology,Wuhan, China. 6MetWare BiotechnologyCo., Ltd.,Wuhan, China. 7Department ofGastrointestinal Surgery, RenminHospital ofWuhanUniversity, Wuhan, China. 8Department of Pancreatic and Gastric Surgery, National Cancer Center, National Clinical Research Center for Cancer, Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. 9Research Center, Beijing Friendship Hospital, Capital Medical University, Beijing, China. 10These authors contributed equally: Liyi Bai, Fayong Hu. e-mail: zhuxuwhu@whu.edu.cn; yibinxie@cicams.ac.cn; zhangshutian@ccmu.edu.cn; minli@ccmu.edu.cn Nature Communications | (2026) 17:6372 1 12 34 56 78 9 0 () :,; 12 34 56 78 9 0 () :,; a non-invasive diagnostic approach for GC screening; however, their practical applicability in clinical settings is significantly limited by the intrinsic limitations between sensitivity and specificity5. Therefore, the development of noninvasive biomarkers with improved sensitivity and specificity remains critically urgent to achieve precise GC screening. Metabolomics has emerged as a promising approach for bio- marker discovery and mechanistic investigations across diseases. Substa