PO.CL01.09 · 临床研究

基于多维cfDNA特征的超灵敏AI驱动MRD检测框架

Ultrasensitive AI-driven framework for MRD detection based on multidimensional cfDNA features

编号 3852 展板 13 时间 4/20 02:00–05:00 区域 Section 45 主讲 Haimeng Tang, MS

分会场 Liquid Biopsies: Circulating Nucleic Acids 3

查看 PDF 下载 PDF 🔒 查看 / 下载完整 PDF 需登录并开通下载套餐 · 查看套餐 / 开通 AACR 官方页面

作者与单位 Authors & Affiliations

Yunfei Shi¹, Hao Zhang², Zexiao Lin³, Ningyou Li², Maolong Wang⁴, Hua Bao², Jinfeng Zhang², Zhili Chang², Yong Ge⁵, Peng Li⁶, Pan Wang⁷, Liang Huang⁸, Xiangming Liu⁹, Lu Han¹⁰, Wangming Ji¹¹, Teng Sun⁹, Dujun Hua², Xunbiao Liu², Mingya Wang², Baihan Zhu², Dongqin Zhu², Xue Wu², Haimeng Tang¹², Hao Zhang⁹, Yang Shao¹³

¹The First Affiliated Hospital of Kunming Medical University, Kunming, China,²Geneseeq Research Institute, Nanjing Geneseeq Technology Inc., Nanjing, China,³Department of Medical Oncology, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China,⁴Department of Thoracic Surgery, Qingdao University Affiliated Hospital, Qingdao, China,⁵Department of Thoracic Surgery, Affiliated Hospital of Xuzhou Medical University, Xuzhou, China,⁶Department of General Surgery, The First Medical Centre, Chinese People’s Liberation Army (PLA) General Hospital, Beijing, China,⁷Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China,⁸Beidahuang Group General Hospital, Branch 1, Harbin, China,⁹Affiliated Hospital of Xuzhou Medical University, Xuzhou, China,¹⁰The First Medical Centre, Chinese People’s Liberation Army (PLA) General Hospital, Beijing, China,¹¹PLA Rocket - Force Characteristic Medical Center, Beijing, China,¹²Geneseeq Technology Inc., Toronto, ON, Canada,¹³Nanjing Geneseeq Technology Inc., Nanjing, China

摘要 Abstract

中文摘要

微小残留病（MRD）检测对于癌症术后风险分层和复发预测至关重要，然而当前的固定panel检测灵敏度和特异性有限，尤其是在肿瘤未知（tumor-naïve）情境下。我们开发了Shielding Ultra，一种针对2,365个癌症相关基因热点突变的泛癌种MRD检测。该检测利用基于唯一分子标识符（UMI）的超深度测序和AI驱动的生物信息学，整合体细胞突变、拷贝数变异（CNV）和片段组学（Frag）特征，在统一工作流程内实现多维MRD评估。分析验证确立了0.0048%的检测限，在晚期术前血浆中展现出94%灵敏度，在健康对照中约99%特异性。肿瘤未知分析在多模态整合后与肿瘤知情工作流程达到98.9%一致性，支持其在肿瘤组织不可获得时的适用性。跨结直肠癌、胆管癌和肺癌队列的临床验证证实了在术后早期时间点的强大预后性能，同时保持高特异性。在结直肠癌中，MRD阳性显示出与复发的稳健关联，风险比高达32.47，在术后监测期间达到高达90.9%的纵向灵敏度。在肺癌中，术后一周的复发检测灵敏度为51.5%，加入术后一个月采样后增至72.7%。跨癌症类型，多模态cfDNA整合增强了检测性能，并实现了在多样临床情境下对MRD的可靠识别。这些发现表明，Shielding Ultra通过整合多维cfDNA特征和基于AI的算法，实现了灵敏且特异的MRD检测。其在多种恶性肿瘤中的强大预后性能以及与肿瘤未知工作流程的兼容性，支持其用于术后风险分层和个性化疾病管理的效用。

查看英文原文 English abstract

Minimal residual disease (MRD) detection is essential for postoperative risk stratification and recurrence prediction in cancer, yet current fixed-panel assays exhibit limited sensitivity and specificity, particularly in tumor-naïve settings. We developed Shielding Ultra, a pan-cancer MRD assay targeting hotspot mutations across 2,365 cancer-related genes. Leveraging ultra-deep unique molecular identifier-based sequencing and AI-driven bioinformatics, the assay integrates somatic mutations, copy number variations (CNVs), and fragmentomic (Frag) features to enable multidimensional MRD assessment within a unified workflow. Analytical validation established a detection limit of 0.0048% and demonstrated 94% sensitivity in late-stage preoperative plasma, with approximately 99% specificity in healthy controls. Tumor-naïve analysis achieved 98.9% concordance with tumor-informed workflows following multimodal integration, supporting applicability when tumor tissue is unavailable. Clinical validation across colorectal, cholangiocarcinoma, and lung cancer cohorts confirmed strong prognostic performance at early postsurgical timepoints while maintaining high specificity. In colorectal cancer, MRD positivity showed a robust association with relapse, yielding hazard ratios up to 32.47 and achieving longitudinal sensitivity of up to 90.9% during postoperative surveillance. In lung cancer, recurrence detection sensitivity was 51.5% at one week after surgery and increased to 72.7% with the addition of one-month postsurgical sampling. Across cancer types, multimodal cfDNA integration strengthened detection performance and enabled reliable identification of MRD across diverse clinical contexts. These findings demonstrate that Shielding Ultra enables sensitive and specific MRD detection through the integration of multidimensional cfDNA features and AI-based algorithms. Its strong prognostic performance across multiple malignancies and compatibility with tumor-naïve workflows support its utility for postoperative risk stratification and personalized disease management.

利益披露 Disclosure

Y. Shi, None.. H. Zhang, None.. Z. Lin, None.. N. Li, None.. M. Wang, None.. H. Bao, None.. J. Zhang, None.. Z. Chang, None.. Y. Ge, None.. P. Li, None.. P. Wang, None.. L. Huang, None.. X. Liu, None.. L. Han, None.. W. Ji, None.. T. Sun, None.. D. Hua, None.. X. Liu, None.. M. Wang, None.. B. Zhu, None.. D. Zhu, None.. X. Wu, None.. H. Tang, None.. H. Zhang, None.

← 返回 AACR 2026 检索