Cell-free DNA Fragmentomics Assay to Discriminate the Malignancy of Breast Nodules and Evaluate Treatment Response

作者：Jiaqi Liu, Yalun Li, Wanxiangfu Tang, Tianyi Qian, Lijun Dai, Ziqi Jia, Heng Cao, Chenghao Li, Yuchen Liu, Yansong Huang, Jiang Wu, Dongxu Ma, Guangdong Qiao, Hua Bao, Shuang Chang, Dongqin Zhu, Shanshan Yang, Xuxiaochen Wu, Xue Wu, Hengyi Xu, Hongyan Chen, Yang Shao, Xiang Wang, Zhihua Liu, Jianzhong Su · 发表于：Genomics Proteomics & Bioinformatics · 年份：2025 · DOI：10.1093/gpbjnl/qzaf028 · 被引用次数：8 · 研究领域：Cancer Genomics and Diagnostics、Genetic factors in colorectal cancer、Molecular Biology Techniques and Applications

The fragmentomics-based cell-free DNA (cfDNA) assays have recently illustrated prominent abilities to identify various cancers from non-conditional healthy controls, while their accuracy for identifying early-stage cancers from benign lesions with inconclusive imaging results remains uncertain. Especially for breast cancer, current imaging-based screening methods suffer from high false positive rates for women with breast nodules, leading to unnecessary biopsies, which add to discomfort and healthcare burden. Here, we enrolled 613 female participants in this multi-center study and demonstrated that cfDNA fragmentomics (cfFrag) is a robust non-invasive biomarker for breast cancer using whole-genome sequencing. Among the multimodal cfFrag profiles, the fragment size ratio (FSR), fragment size distribution (FSD), and copy number variation (CNV) show more distinguishing ability than Griffin, motif breakpoint (MBP), and neomer. The cfFrag model using the optimal three fragmentomics features discriminated early-stage breast cancer from benign nodules, even at a low sequencing depth (3×). Notably, it demonstrated a specificity of 94.1% in asymptomatic healthy women at a 90% sensitivity for breast cancer. Moreover, we comprehensively showcased the clinical utility of the cfFrag model in predicting patient responses to neoadjuvant chemotherapy (NAC) and its enhanced performance when combined with multimodal features, including radiological results [area under the curve (AUC) = 0.93-0....