Liquid Biopsy Measures Epigenetic Instability in Cancer

Liquid Biopsy, Epigenetic Instability, DNA methylation, Early cancer detection, Oncology diagnostics, Cell-free DNA, Cancer screening, Precision oncology, Molecular diagnostics, early cancer detection, DNA methylation variability, cell-free DNA testing, oncology diagnostics, Epigenetic Instability Index, cancer screening biomarkers, blood-based cancer test, precision oncology, lung cancer detection, breast cancer screening, machine learning in oncology, cancer epigenomics, molecular cancer diagnostics

Key Takeaways

Johns Hopkins researchers developed a novel liquid biopsy that detects early-stage cancers by measuring epigenetic instability rather than absolute DNA methylation levels.
A new metric, the Epigenetic Instability Index (EII), showed strong accuracy in identifying early lung and breast cancers from blood samples.
The approach may overcome population bias seen in existing methylation-based liquid biopsies and could support earlier, more reliable cancer screening.

Rethinking Liquid Biopsy Through Epigenetic Instability

Early cancer detection remains a critical unmet need in oncology, particularly for tumors that shed limited DNA into the bloodstream. Researchers at the Johns Hopkins Kimmel Cancer Center now report a promising liquid biopsy strategy that focuses on random variation in DNA methylation, termed epigenetic instability, rather than fixed methylation changes.

Explore All Oncology CME Conferences 2026

Published in Clinical Cancer Research, the proof-of-concept study introduces the Epigenetic Instability Index (EII), a quantitative measure of methylation stochasticity in circulating cell-free DNA. According to the investigators, this randomness appears to be a consistent hallmark of early carcinogenesis across multiple cancer types.

How the Epigenetic Instability Index in Liquid Biopsy Works?

Traditional methylation-based liquid biopsies rely on detecting specific methylation patterns, which often vary by age, ethnicity, or cohort characteristics. To overcome this limitation, the Johns Hopkins team analyzed 2,084 publicly available cancer methylation samples, identifying 269 CpG islands that captured the highest methylation variability across cancers.

Using these regions, researchers trained a machine learning model to differentiate cancer-derived signals from healthy samples. The EII demonstrated strong diagnostic performance:

Stage 1A lung adenocarcinoma detected with 81% sensitivity at 95% specificity
Early-stage breast cancer detected with ~68% sensitivity at 95% specificity
Detectable signals were also observed in colon, brain, pancreatic, and prostate cancers

For clinicians, this suggests a potentially more universal and population-agnostic biomarker for early cancer detection.

Clinical Relevance and Future Use

From a clinical perspective, EII could serve as a secondary triaging tool alongside established screening tests. For example, patients with elevated PSA levels, often associated with false positives, could benefit from an EII-based blood test to guide biopsy decisions more accurately.

The research team is now expanding validation studies in larger, longitudinal cohorts. If confirmed, EII may complement other Johns Hopkins-developed assays such as DELFI and mutation-based liquid biopsies, supporting earlier interception of cancer development.

To know more about emerging models in integrated cancer and cardiology care, read the full OCC2026 blog and register on eMedEvents.

For oncologists, pathologists, nurses, and preventive care specialists, this approach highlights a shift toward epigenome-wide variability metrics as clinically meaningful cancer signals.