Johns Hopkins’ liquid biopsy method offers promising cancer detection tool

A group of scientists from the Johns Hopkins Kimmel Cancer Center have discovered a new method for diagnosing lung and breast cancers early, according to an announcement. The pathway is liquid biopsy based, and measures “the random variation in DNA methylation patterns.”

The technique leverages the Epigenetic Instability Index (EII) and machine learning. The researchers reviewed “publicly available cancer DNA methylation datasets from 2,084 samples” and pinpointed 269 CpG islands that “captured most DNA methylation variability across multiple cancer types.” They then utilized machine learning to differentiate between cancer and healthy signals. Stage 1A lung adenocarcinoma was detected with 81% sensitivity at 95% specificity, while early-stage breast cancer was found with 68% sensitivity at 95% specificity, according to Johns Hopkins.

The team is working to enhance the method for clinical settings.