Hepta’s liquid biopsy-native AI model links coordinated methylation and gene expression changes to fibrosis progression, showing that liver biology can be measured non-invasively from blood

Hepta, a biotechnology company using transformer-based AI to decode the cell-free DNA (cfDNA) epigenome and detect organ-specific signals of chronic disease, today announced the creation of an unparalleled multi-omic atlas of metabolic dysfunction–associated steatohepatitis (MASH) in collaboration with Duke University. This work builds on Hepta’s expanded cfDNA studies with Mainz University and Akero Therapeutics’ Phase 3 clinical trial program, further strengthening the company’s partnerships across academic and clinical research.

The atlas integrates single-cell and bulk data across hundreds of liver samples, including gene expression, chromatin accessibility, and DNA methylation, paired with cfDNA methylation from matched blood plasma. Presented this week at AASLD’s The Liver Meeting, the atlas reveals how coordinated methylation and transcriptional programs drive inflammation and fibrosis and demonstrates that these same molecular signals are detectable in blood using Hepta’s liquid biopsy-native AI platform.

“In liver tissue, we observe pathway activity, gene expression, and cell type composition that are mediated by methylation across disease severity,” said Anna Mae Diehl, M.D., Florence McAlister Distinguished Professor of Medicine at Duke University. "Critically, those same signatures are measurable in plasma cfDNA, indicating that cfDNA methylation can serve as a faithful mirror of liver biology. That concordance supports cfDNA methylation as a plausible, non-invasive readout of pathway activities in liver disease, offering insights far deeper than traditional biomarkers.”

The MASH atlas maps how methylation governs pathway-level activity across hepatocytes, cholangiocytes, and stromal cells in the liver. The analysis shows the role of methylation in a broad dysregulation of metabolic, inflammatory, fibrogenic, and bile-acid pathways that track with fibrosis severity, including mechanisms targeted by emerging therapeutic classes such as GLP-1, FGF-21, and THR-β agonists. The MASH atlas also demonstrates that the same pathway signatures observed in liver tissue are reproducibly detectable in cfDNA circulating in blood, creating a molecular bridge between liver-tissue biology and liquid-biopsy measurement.

This foundational work, in addition to Hepta’s expanded cfDNA results spanning across multiple independent cohorts — including Duke University, Mainz University, and Akero Therapeutics’ phase 3 clinical trial program — was featured in Dr. Jörn Schattenberg’s presentation at the Precision Liver Symposium.

“Across independent cohorts, we see strong, consistent diagnostic performance from Hepta’s cfDNA methylation platform,” said Jörn Schattenberg, M.D., Director of the Metabolic Liver Research Center at Saarland University. “What’s striking is that the same readout also delivers deep biological insights. It’s not just a yes-or-no test; these signals reveal how repair and fibrosis programs evolve, which opens the door to earlier intervention and therapy guidance.”

Hepta’s transformer-based, liquid biopsy–native AI was specifically designed to resolve the complexity of cfDNA. The model interprets billions of cfDNA fragments in context, detecting subtle, distributed patterns that correspond to tissue-level changes. By decoding the entire cfDNA epigenetic landscape, the platform enables tissue-level biological interpretation from a standard blood draw.

While traditional panel biomarkers infer fibrosis indirectly, cfDNA methylation directly reads the gene-regulatory programs driving those changes. Together with the multi-omic atlas, these data represent the largest multi-cohort validation to date of cfDNA methylation as a biomarker in MASH and demonstrate how deep multi-omics can inform clinical translation.

“Our data demonstrate that cfDNA captures the molecular fingerprint of disrupted repair,” said Hamed Amini, Ph.D., Co-founder and CEO of Hepta. “The atlas establishes the biological foundation for our technology, showing that the same pathways we observe in tissue are encoded in blood. This work lays a foundation for biology-driven detection and for future therapeutic strategies that address the mechanisms behind liver damage and fibrosis.”

Hepta’s approach extends the frontier of liquid biopsy beyond oncology, positioning cfDNA methylation as a scalable molecular window into chronic disease biology and, potentially, a snapshot of full body health. The company is expanding its clinical studies to further validate cfDNA-based biomarkers for early detection, informing treatment strategies, and longitudinal monitoring across the care continuum.

About Hepta

Hepta detects chronic disease early by using cfDNA epigenetic analysis and AI to deliver accurate, non-invasive diagnostics at population scale. Founded by former Illumina, Grail and Google scientists, the company is backed by Felicis Ventures, Illumina Ventures, and SeaX Ventures. Hepta is proving that blood-based epigenetic biomarkers can replace invasive biopsies, enabling earlier and more accessible detection of diseases like MASH and laying the foundation for future applications across chronic diseases. For more information, visit www.hepta.bio.

View source version on businesswire.com: https://www.businesswire.com/news/home/20251110261207/en/