Particle.news
Download on the App Store

AI Fragmentome Blood Test Improves Detection of Liver Cancer

A July 31 study finds that combining genome-wide cfDNA fragmentome scores with AFP, with simple clinical risk factors, raises sensitivity for early hepatocellular carcinoma.

Overview

  • The DELFI fragmentome classifier was validated in 377 patients from Guatemala and Romania and, when combined with alpha-fetoprotein (AFP) and basic risk factors, detected hepatocellular carcinoma more often than AFP alone, including early-stage tumors.
  • Researchers introduced MethID, a methylation-based method that traced circulating DNA back to hepatocytes, blood-vessel cells, and immune cells and showed those signals help the test identify liver cancer.
  • The genome-wide analysis found a circulating DNA mutation pattern linked to aflatoxin exposure in Guatemalan participants, marking the first identification of that signature across the whole genome in plasma.
  • The test uses less than 1 mL of plasma and low-coverage whole-genome sequencing to read fragmentation, methylation, mutations, chromatin signals, and repetitive elements, and higher fragmentome scores correlated with tumor features and survival.
  • Authors and funders say the results point to a scalable, multimodal liquid-biopsy approach but stress that prospective clinical trials, cost-effectiveness studies, and implementation work are needed before routine screening or monitoring changes for patients.