Abstract:Abstract Background: Hepatocellular carcinoma (HCC) is the most common form of liver cancer and accounts for ∼90% of cases globally. Detection of early-stage HCC is vital as it allows for potentially curative treatment, yet remains a significant challenge. Micronuclei (MN) are a hallmark of genomic instability. An increased frequency of MN is commonly observed in tumor cells as well as other somatic cells, including erythrocytes, in cancer patients. Here, we have established a technique (WO2021/228246 A1), for the isolation and analysis of micronuclei DNA (MN-DNA) in erythrocytes from peripheral blood. Significant changes in read densities at specific genomic locations within MN-DNA were identified in patients, termed as tumor- associated MN-DNA (taMN-DNA) features. This study evaluates the potential of these taMN- DNA features for early-stage HCC detection across human and murine model. Methods: To explore the potential of MN-DNA in cancer detection, we first collected 1-2 mL of whole blood from HDs (N = 53) and HCC patients (N = 53). MN-DNA was isolated and purified from erythrocytes, followed by whole-genome sequencing. Participants were randomly divided into training, test cohorts in an 8:2 ratio. We applied machine learning algorithms to leverage these features for cancer detection in the human model. Additionally, to investigate the formation of taMN-DNA features, we employed a well-established mouse model that mimics HCC development. Mice were injected with the genotoxic agent diethylnitrosamine (DEN) 2 weeks after birth, and developed malignant macroscopic liver nodules at approximately 40 weeks of age. Results: In the human cohort, we enrolled 106 participants, with more than half of the HCC cases were diagnosed at early-stage (stage 0-II). The cancer detection model utilizing taMN- DNA features achieved an overall accuracy of 86.3%, with a sensitivity of 81.8% and specificity of 90.9%. For early-stage, the model demonstrated sensitivities of 54.5% at a specificity of 90.9%. In the DEN-induced HCC mouse model, unsupervised hierarchical clustering based on these selected taMN-DNA features clearly separated WT and tumor mice into two distinct groups. Notably, mice that were not tumorigenic at 36 weeks post-treatment in the DEN-treated group fell into the WT group when classified by the same taMN-DNA features, indicating that the formation of taMN-DNA signatures is associated with the presence of tumors specifically. Conclusions: This pilot study highlights the potential of MN-DNA as a promising tool for early HCC detection. Leveraging these taMN-DNA features can accurately distinguish early-stage HCC patients from HDs with high sensitivity. In both human and murine models, there is a significant relationship between these signatures and tumor presence, suggesting that taMN- DNA features might reflect the diseased state across mammalian species. Research sponsor: Timing Biotech. Citation Format: Hui Lin, Haobo Sun, Xingyun Yao, Xiaoxiao Fan, Fei Meng, Honghao Liang, Xiaofei Gao. Leveraging micronuclei DNA from erythrocytes for early detection of hepatocellular carcinoma [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr B073.

Machine learning-based genome-wide interrogation of somatic copy number aberrations in circulating tumor DNA for early detection of hepatocellular carcinoma.

Abstract B073: Leveraging Micronuclei DNA from Erythrocytes for Early Detection of Hepatocellular Carcinoma

Dynamical Detection and Characterization of Circulating Tumor DNA in Patients with Hepatocellular Carcinoma

Early detection of hepatocellular carcinoma using cfDNA signatures from cirrhotic patients with nodules.

Genome-wide Plasma Cell-Free DNA Methylation Profiling to Identify High-Performing Biomarkers for Early Detection of Hepatocellular Carcinoma.

Genome-wide Mapping of 5-Hydroxymethylcytosines in Circulating Cell-Free DNA As a Non-Invasive Approach for Early Detection of Hepatocellular Carcinoma

Longitudinal and Personalized Detection of Circulating Tumor DNA (ctdna) for Monitoring Efficacy of Atezolizumab Plus Bevacizumab in Patients with Unresectable Hepatocellular Carcinoma (HCC).

Liquid Biopsy by Combining 5-Hydroxymethylcytosine Signatures of Plasma Cell-Free DNA and Protein Biomarkers for Diagnosis and Prognosis of Hepatocellular Carcinoma

A Computational Framework To Simultaneously Quantify Dna Methylation, Somatic Copy Number Alternation And Dna Heterogeneity From Low Coverage Plasma Circulating Dna Sequencing

Genome-Scale Profiling of Circulating Cell-Free DNA Signatures for Early Detection of Hepatocellular Carcinoma in Patients with Cirrhosis

Detection and monitoring of HBV-related hepatocellular carcinoma from plasma cfDNA fragmentation profiles

Development and validation of a non-invasive cfDNA targeted sequencing assay for early-stage hepatocellular carcinoma detection using cfDNA methylation and fragmentomics.

Early cancer detection using low-coverage whole-genome sequencing of cell-free DNA fragments.

Early Detection of Hepatocellular Carcinoma with Methylation and Fragmentation Signatures of Circulating Tumour DNA: a Prospective, Multicentre, Case-Control, Observational Study

Abstract 782: the Genetic and Epigenetic Abnormalities of Plasma Cfdna As Liquid Biopsy Biomarkers to Diagnose Hepatocellular Carcinoma

Discovery and clinical validation of cost-effective noninvasive early detection of hepatocellular carcinoma (HCC) through circulating tumor DNA (ctDNA) methylation signature.

Comprehensive Liquid Profiling of Circulating Tumor DNA and Protein Biomarkers in Long-Term Follow-Up Patients with Hepatocellular Carcinoma

Noninvasive Detection of Hepatocellular Carcinoma with Circulating Tumor DNA Features and Α-Fetoprotein

Machine Learning-Enabled Early Detection of Hepatocellular Carcinoma Utilizing Cell-Free DNA Mutation and Fragmentation Multiplicity: a Prospective Study

Ultrasensitive and affordable assay for early detection of primary liver cancer using plasma cell‐free DNA fragmentomics

A computational framework to unify orthogonal information in DNA methylation and copy number aberrations in cell-free DNA for early cancer detection