AI Blood-Based Lung Cancer Screening Test Developed for Fragmentome

MONDAY, June 10, 2024 (HealthDay News) -- A novel blood-based lung cancer screening test has been developed and validated using genome-wide sequencing to analyze cell-free DNA (cfDNA) fragmentation profiles, according to a study published online June 3 in Cancer Discovery.

Noting that changes in genome-wide cfDNA fragmentation profiles (fragmentomes) in peripheral blood reflect genomic and chromatin characteristics of lung cancer, Peter J. Mazzone, M.D., M.P.H., from the Cleveland Clinic, and colleagues conducted a prospective case-control study among 958 individuals eligible for lung cancer screening to develop a blood-based lung cancer detection test, which is followed by low-dose computed tomography when positive. To identify individuals who were more or less likely to have lung cancer, machine learning was applied to fragmentome features. The classifier was trained using 576 cases and controls from study samples and validated in a held-out group of 382 cases and controls.

The researchers found that high sensitivity for lung cancer was demonstrated in the validation, which was consistent across demographic groups and comorbid conditions. There was potential to prevent thousands of lung cancer deaths by applying test performance to the screening-eligible population in a five-year model with modest utilization assumptions.

"While we await further validation in ongoing prospective cohort studies, modeling suggests substantial public health benefits if a test like this can improve lung cancer screening participation among those who are not currently receiving it," the authors write.

The study was supported in part by DELFI Diagnostics; two authors are inventors of patent applications related to cfDNA for cancer detection.

Abstract/Full Text (subscription or payment may be required)