Blood-based DNA methylation marker model for short-term and long-term lung cancer risk prediction.
Summary
A novel study developed and validated a blood-based DNA methylation marker model (BBDMM) for predicting lung cancer risk. This model utilizes informative CpG sites identified through epigenome-wide association studies (EWAS). It was developed and internally validated in a German cohort of over 2,400 participants, then externally validated in Norwegian cohorts. The BBDMM demonstrated excellent discriminatory capacity, with AUCs of 0.84 and 0.85, and predictive stability over periods up to 18 years before diagnosis. These findings suggest significant potential for improving lung cancer screening.
Analysis
🔴 CLINIQUE: This BBDMM model represents a promising prognostic biomarker for lung cancer risk, with AUCs ranging from 0.84 to 0.85, indicating strong discriminatory capacity. It could enable more precise identification of high-risk populations, including never smokers, for more targeted and cost-efficient screening than low-dose computed tomography (LDCT) alone. Larger prospective validation could justify a change in clinical practice, integrating this blood test to stratify risk and optimize screening programs. The time to clinical impact could be 3-5 years. 🟢 BIOMOL: The discovery lies in identifying specific CpG sites whose DNA methylation is informative for lung cancer risk, through epigenome-wide association studies (EWAS). The approach uses a liquid biopsy (blood) to detect these epigenetic markers, which is less invasive and more accessible than tissue biopsies. Analytical validity is suggested by the reproducibility of AUCs across different cohorts and over varying follow-up periods. 🔵 BIOINFO: The model was developed and validated on large cohorts (ESTHER with 2,459 participants, HUNT2/HUNT3 with 233 participants). The model's architecture is based on informative CpG sites, and its performance was evaluated using AUC, demonstrating stable and robust discrimination. External validation on independent cohorts and over both short-term and long-term follow-ups strengthens the model's reproducibility and generalizability.