Synergizing Instant Digital Pathology and Artificial Intelligence: Towards Near-Real-Time Molecular Profiling in Lung Cancer

Introduction: Non-small-cell lung cancer (NSCLC) is a leading cause of death, frequently diagnosed at advanced stages. Current guidelines emphasize the necessity of molecular testing to identify driver mutations for targeted therapies. This study aims to develop a rapid diagnostic workflow combining instant digital microscopy and artificial intelligence (AI) to predict the most likely driver mutation based on tissue morphology and clinical data. Materials and methods: Fresh samples from EBUS/TBNA obtained from lung cancer patients were analyzed with VivaScope2500, a confocal laser microscope providing real-time native digital tissue images, without sample damage. Pathologists ensured tissue adequacy and manually annotated images. Images were used to train an AI model to predict the most likely driver mutation. The predicted KRAS, EGFR and BRAF mutations were rapidly tested with Idylla (Biocartis); others underwent NGS. AI predictions were compared with molecular profile obtained from paired paraffin sections. Result: The AI model demonstrated good performance in identifying KRAS, EGFR and BRAF mutations from digital images. Compared to the ground truth, the algorithm achieved 78,6% accuracy in guiding the selection of the most appropriate molecular test (PCR or NGS). Validation on a broader, independent cohort is currently ongoing. Conclusion: This study demonstrated the feasibility and value of combining instant digital microscopy and AI in assisting pathologists in NSCLC diagnosis and molecular test selection. The rapid workflow enables faster clinical decisions and sustains accessibility to targeted treatments, potentially improving diagnostic accuracy and patient outcomes in clinical practice. Funded by PNRR-POC-2022-12376531.