WCLC 2025 - Posters & ePosters-P1.04.48 Rapid Intra-Operative Detection of Visceral Pleural Invasion in Non-Small Cell Lung Cancer Using Label-Free Multiphoton Microscopy

P1.04.48 Rapid Intra-Operative Detection of Visceral Pleural Invasion in Non-Small Cell Lung Cancer Using Label-Free Multiphoton Microscopy

Back to course

Pdf Summary

This study presents a novel approach using label-free multiphoton microscopy (MPM) for rapid intraoperative detection of visceral pleural invasion (VPI) in non-small cell lung cancer (NSCLC). VPI is a critical pathological factor influencing prognosis and surgical decisions, but current detection methods are limited and largely postoperative.<br /><br />The researchers developed a portable MPM system capable of visualizing the elastic layer and collagen architecture of the visceral pleura in real time without staining, producing images comparable to the gold-standard elastin van Gieson (EVG) histological staining. The study involved three blinded pathologists who assessed MPM images against conventional histology, demonstrating that MPM achieved a diagnostic accuracy of 96.4%, surpassing traditional methods at 95.3%, with high inter-observer consistency.<br /><br />Additionally, the team created a machine learning classifier based on 28 quantitative collagen biomarkers derived from MPM images. This automated system showed excellent performance in grading VPI, achieving a macro-average area under the curve (AUC) of 0.88, highlighting its potential to facilitate consistent, objective diagnosis.<br /><br />The key clinical impact is shifting VPI diagnosis from postoperative to intraoperative, enabling surgeons to make immediate, personalized decisions regarding resection extent during NSCLC surgery. This advancement promises improved surgical precision and potentially better patient outcomes.<br /><br />Future directions include large-scale multicenter clinical trials to validate MPM's diagnostic robustness and long-term patient correlations. The development of miniaturized MPM probes integrated with thoracoscopic systems is also proposed to allow in vivo, real-time imaging during surgery.<br /><br />In summary, this study establishes label-free MPM combined with machine learning as an accurate, rapid, and practical tool for intraoperative VPI detection, offering a significant step forward in personalized lung cancer surgery.

Asset Subtitle

Suyi Chen

Meta Tag

Speaker Suyi Chen

Topic Screening and Early Detection

Keywords

label-free multiphoton microscopy

visceral pleural invasion

non-small cell lung cancer

intraoperative detection

collagen biomarkers

machine learning classifier

diagnostic accuracy

real-time imaging

thoracoscopic integration

personalized lung cancer surgery