Abstract B020: Early detection of colorectal cancer using micronuclei DNA
Abstract Background: Current development in stool- and cell-free DNA (cfDNA)-based technologies have demonstrated promising potentials in colorectal cancer (CRC) diagnosis. However, the early detection of CRC and advanced adenomas (AA) remains a challenge. Micronuclei (MN) are hallmarks of genomic instabilities. Elevated levels of MN+ erythrocytes have been studied in genotoxicity testing and cancer diagnosis. We utilized previously developed methods to isolate and sequence MN-DNA from erythrocytes in 1-2 mL peripheral blood, revealing significant differences between patients and healthy donors (WO2021/228246 A1). Here, we established an independent cohort to further evaluate the value of MN-DNA in identifying CRC and AA. Methods: We enrolled 1251 samples with complete colonoscopy reports, comprising 360 CRC patients, 330 AA patients, and 561 non-cancer individuals. We collected 1-2 mL of peripheral blood from each enrolled individual, and the MN-DNA was isolated and performed on whole genome sequencing. We used 80% of these samples for the CRC- and AA-associated MN-DNA features identification and model development, while the remaining 20% were designated as the test set to evaluate the model performance and determine the cutoff. Based on the specificity and sensitivity in the test set, we launched a prospective observational case-control clinical trial (NCT05875584), which was used as an independent cohort to validate the model and to perform a comparative analysis with quantitative fecal immunochemical test (qFIT). A total of 598 samples were enrolled in this study (585 were available, including 299 HDs, 206 AAs, and 80 CRCs). Results: The predictive CRC model built on MN-DNA features achieved an AUC of 96% and the detection of AAachieved an AUC of 88% in the independent cohort. Comparing MN-DNA with qFIT, the overall sensitivity was generally consistent (90% vs. 88%), with MN-DNA achieving 91% sensitivity in early-stage (stage I) CRC, significantly higher than 68% in qFIT. Additionally, MN-DNA showed superior sensitivity in detecting AA (70% vs. 18%), with comparable specificity (90% vs. 96%). Conclusions: Our results demonstrate that MN-DNA within erythrocytes from 1-2ml peripheral blood enables accurate detection of AA and early-stage CRC. Research sponsor: Timing Biotech. Citation Format: Xingyun Yao, Haobo Sun, Yurong Jiao, Chengcheng Liu, Xiangxing Kong, Honghao Liang, Fei Meng, Jun Li, Kefeng Ding, Xiaofei Gao. Early detection of colorectal cancer using micronuclei DNA [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Functional and Genomic Precision Medicine in Cancer: Different Perspectives, Common Goals; 2025 Mar 11-13; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2025;85(5 Suppl):Abstract nr B020.
Authors
- Yurong Jiao
- Xingyun Yao (ORCID: https://orcid.org/0000-0002-9795-9722)
- Xiaofei Gao (ORCID: https://orcid.org/0000-0003-2799-7421)
- Kefeng Ding (ORCID: https://orcid.org/0000-0002-2380-3717)
- Chengcheng Liu (ORCID: https://orcid.org/0000-0002-5573-6467)
- Honghao Liang
- Haobo Sun (ORCID: https://orcid.org/0000-0002-7152-5731)
- Xiangxing Kong (ORCID: https://orcid.org/0000-0001-8744-5421)
- Fei Meng
- Jun Li (ORCID: https://orcid.org/0009-0004-1683-5130)
Institutions
- Westlake University (CN)
- Second Affiliated Hospital of Zhejiang University (CN)
Publication Details
- Journal
- Cancer Research
- Published
- 2025-03-11
- DOI
- https://doi.org/10.1158/1538-7445.genfunc25-b020
- Primary Topic
- Cancer Genomics and Diagnostics
- Type
- article
- Field-Weighted Citation Impact
- 0.00