Blood Test Detects Early-Stage Cancers by Measuring Epigenetic Instability

Many liquid biopsies rely on measuring absolute DNA methylation changes, but these signals can vary widely across populations and lose accuracy outside narrowly defined cohorts. Researchers have now shown that focusing instead on random variation in DNA methylation — a feature of early cancer development — can more reliably distinguish early cancers from healthy tissue. This strategy enables highly accurate detection of early-stage disease from a simple blood test.

Researchers at Johns Hopkins Kimmel Cancer Center (Baltimore, MD, USA) have developed a new liquid biopsy framework based on a metric called the Epigenetic Instability Index (EII), which quantifies stochastic variation in DNA methylation rather than absolute methylation levels. This approach is designed to be more universal and less dependent on population-specific reference profiles, addressing a key limitation of current methylation-based diagnostics.

To develop the method, the team analyzed publicly available DNA methylation data from 2,084 cancer samples spanning multiple tumor types. From these datasets, they identified 269 CpG islands that capture the majority of methylation variability across cancers. These regions served as the basis for training a machine learning model to distinguish cancer-derived cell-free DNA signals from background variation in blood samples.

When applied to blood-based data, the EII approach demonstrated strong performance in detecting early-stage cancers. In lung adenocarcinoma, the method identified stage IA disease with 81% sensitivity at 95% specificity. For early-stage breast cancer, the test achieved approximately 68% sensitivity at the same specificity threshold, according to findings published in Clinical Cancer Research.

In addition to lung and breast cancer, the EII approach also showed promise in detecting early signals from colon, brain, pancreatic, and prostate cancers. By capturing epigenetic instability that emerges at the earliest stages of tumorigenesis, the method could complement existing mutation-based liquid biopsies. Researchers are now working to refine the technique and validate it in larger, long-term clinical cohorts, to integrate EII into routine cancer screening and risk stratification.

“This is the first study where we are trying to really implement measuring that variation, or stochasticity, into a diagnostic tool,” said lead study author Hariharan Easwaran, Ph.D., M.Sc. “Our hypothesis is that during the earliest stages of cancer development, methylation starts shifting. We can try to pick those signals using these stochasticity metrics, even of early cancer stages, as long as the DNA is shed in the blood.”

Related Links:
Johns Hopkins Kimmel Cancer Center