Abstract
The early detection and diagnosis of lung cancer through extracellular vesicle (EV)-based liquid biopsy show substantial promise for enhancing clinical outcomes. Nonetheless, there is a scarcity of large-scale clinical investigations validating EV-based liquid biopsy. To evaluate the EV membrane protein panel as a diagnostic tool for early-stage cancer detection and validate its efficacy and clinical applicability, a cohort comprised of 302 individuals without cancer and 645 with lung cancer was recruited. Participants were randomly divided into training and validation cohorts at a 1:1 ratio while maintaining the proportion of different subtypes. A diagnostic panel (EV early lung cancer membrane protein 5, EVELC-M5) consisting of five EV membrane proteins (CD81, PDL1, GLIPR1, LBR and SFTPA1) was developed using a High-throughput Nano-biochip Integrated System for Liquid Biopsy (HNCIB) to realize rapid analysis of a large cohort of patient samples at a single EV level. EVELC-M5 could accurately differentiate patients with early lung cancer from the control group. The area under the curve (AUC) of EVELC-M5 for distinguishing patients with early lung cancer from the control group in the validation cohort was 0.926, and the AUC for diagnosing patients with early lung cancer with lung nodules ≤ 8 mm was 0.931. EV-SFTPA1 proved to be the most effective marker, exhibiting a sensitivity of 89.4% in patients with early lung cancer. To our knowledge, this is the first study to use EV-SFTPA1 for early lung cancer detection, elucidating its robust tissue specificity. Collectively, the findings highlight that EVELC-M5 in conjunction with HNCIB is an effective diagnostic toolset for detecting early lung cancer and substantially promotes its diagnosis. Trial Registration: ClinicalTrials.gov identifier: ChiCTR2300072317.