EUS-Anchored Multimodal Evaluation of Pancreatic Cystic Lesions: Toward a Conceptual Diagnostic Framework

Pancreatic cystic lesions (PCLs) represent a growing clinical challenge due to their diverse biological behaviors and the substantial overlap in imaging features between benign, premalignant, and malignant entities. Traditional diagnostic approaches relying on cross-sectional imaging or isolated morphologic criteria frequently fail to achieve adequate risk discrimination. Advances in endoscopic ultrasound (EUS) now permit detailed morphologic assessment complemented by cyst-fluid biochemical markers, proteomic signatures, and comprehensive genomic profiling using next-generation sequencing. Parallel progress in artificial intelligence (AI) further strengthens diagnostic precision by integrating EUS features with multimodal biomarker data to reduce subjectivity and support individualized clinical decision-making. This review introduces an EUS-based multimodal diagnostic framework of PCLs that integrates morphological evaluation, cyst-fluid biochemical testing, molecular profiling, and AI-assisted analysis. By synthesizing current evidence, we outline how the integrative approach enhances diagnostic accuracy, biological interpretability, and individualized risk stratification for PCLs.