Evaluating the Use of Circulating MicroRNA Profiles for Lung Cancer Detection in Symptomatic Patients

Area of Science:

• Oncology
• Molecular Biology
• Biomarker Discovery

Background:

• Lung cancer has a low survival rate, necessitating improved early detection methods.
• MicroRNA (miRNA) signatures in blood show potential as sensitive and specific biomarkers.
• Further validation in large cohorts is required to confirm the generalizability of miRNA signatures.

Purpose of the Study:

• To evaluate blood-borne miRNAs as circulating markers for lung cancer detection.
• To assess the sensitivity and specificity of miRNA signatures in a large patient cohort.

Main Methods:

• A multicenter cohort study including 3102 patients (March 2009 - March 2018).
• Genome-wide miRNA profiles from 3046 blood samples analyzed using machine learning.
• Three classification scenarios tested: lung cancer vs. all others, lung cancer vs. non-tumor lung disease, and early-stage lung cancer vs. no lung cancer.

Main Results:

• A 15-miRNA signature distinguished lung cancer from all others with 91.4% accuracy, 82.8% sensitivity, and 93.5% specificity.
• A 14-miRNA signature differentiated lung cancer from non-tumor lung diseases with 92.5% accuracy, 96.4% sensitivity, and 88.6% specificity.
• A 14-miRNA signature identified early-stage lung cancer with 95.9% accuracy, 76.3% sensitivity, and 97.5% specificity.

Conclusions:

• Identified miRNA patterns show potential as a component of a minimally invasive lung cancer diagnostic test.
• miRNA signatures can complement existing diagnostic tools like imaging and biopsy.
• This approach may improve lung cancer detection rates and patient outcomes.