Serial EUS-Guided FNA for the Surveillance of Pancreatic Cysts: A Study of Long-Term Performance of Tumor Markers

Background and Aim The natural history of KRAS mutations in mucinous pancreatic cysts (MPCs) over time remains to be fully understood. The aim of this study was to examine the performance of DNA markers and assess changes of KRAS mutations over time. Methods Patients who underwent EUS-FNA of pancreatic cysts with at least two separate molecular analysis results were included in the study. We assessed the baseline patient and cyst characteristics, and DNA fluid analysis. The presence of either a KRAS mutation, or a CEA > 192 ng/ml was used as the diagnostic standard for mucinous cysts when surgical pathology was not available. Results A total of 933 pancreatic cyst fluid samples were collected, including 117 with >= 2 FNAs. Examinations were performed over a median of 30 months (range 1-115 months). Forty-three (36%) had a mutant KRAS on the index analysis out of which 26 had a change in their KRAS status to the wild-type. Eighty-one (64%) had a wild-type KRAS on the index analysis out of which 18 had change in their KRAS status to mutant type. There was no significant difference in the index cyst characteristics, presence of symptoms, or main duct involvement based on KRAS status change. Increasing age was associated with a changing KRAS mutation status (p = 0.023). Conclusion KRAS mutations gain and loss in pancreatic cyst fluid appears to occur frequently during long-term surveillance of MPCs. Age appears to be the only predictor for KRAS change over time.

Automated summary

This study aims to examine the natural history of KRAS mutations in mucinous pancreatic cysts (MPCs) and found that KRAS mutations in pancreatic cyst fluid frequently occur during long-term surveillance, with age being the only predictor for KRAS changes over time.