Single GNAS Droplet-Based Digital Polymerase Chain Reaction Analysis of Pancreatic Cyst Fluid: An Effective Up-Front Strategy for Mucinous Cyst Diagnosis by Endoscopic Ultrasound-Guided Fine-Needle Aspiration.

INTRODUCTION

Accurate diagnosis of mucinous pancreatic cystic neoplasms (mPCNs) remains a clinical challenge. This study investigated the utility of single GNAS droplet-based digital polymerase chain reaction (ddPCR) analysis as a novel approach to refine the diagnostic accuracy of mPCNs using endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA).

METHODS

Patients who underwent EUS-FNA and GNAS pancreatic cyst fluid (PCF) analyses for pancreatic cystic lesion (PCL) assessment were prospectively enrolled. Cysts were categorized as mPCNs, non-mPCNs, or inconclusive PCLs (iPCLs) by integrating increasing information levels: high-resolution imaging and non-DNA PCF features (level 1), GNAS PCF analysis (level 2), and surgical pathology (level 3).

RESULTS

One hundred forty patients were included, 25 of whom underwent pancreatic surgery. Level 1 identified 68 mPCNs (49%), 24 non-mPCNs (17%), and 48 iPCLs (34%). GNAS mutations were detected in 42 of 68 (62%) mPCNs, 1 of 24 (4%) non-mPCNs, and 16 of 48 (33%) iPCLs. Level 2 increased mPCN detection to 62% and reduced iPCLs by one-third. Mutated GNAS showed 66% sensitivity for diagnosing mPCNs in the whole cohort and 65% in resected cases, outperforming both imaging and non-DNA PCF mucinous criteria, with 100% specificity and limited concordance with carcinoembryonic antigen, cytology, and fluid viscosity, highlighting its complementary diagnostic value. Cost-effectiveness simulations for iPCLs demonstrated that GNAS -ddPCR significantly reduced diagnostic costs by 24% compared with next-generation sequencing testing.

DISCUSSION: Single GNAS- ddPCR analysis in PCF supported mPCNs diagnosis in 62% of cases and uncovered 33% of iPCLs as mPCNs with 100% specificity. It adds complementary value to standard cyst fluid markers offering a simple and cost-effective tool for improving PCL diagnosis by EUS-FNA.