Immunostaining as an adjunct to cytology for diagnosis of pancreatic adenocarcinoma

Abstract

Background & aims: Serial analysis of gene expression has helped identify several proteins that are expressed differentially in pancreatic cancer and are highly sensitive and specific for pancreatic adenocarcinoma. We evaluated if the diagnostic accuracy of pancreatic cancer from endoscopic ultrasound (EUS)-fine-needle aspiration (FNA) can be improved by combined evaluation of cytology and immunostaining with these markers.

Methods: This study involved the use of archived specimens from patients treated at Saint Louis University Hospital from 2002 to 2006. We identified 5 protein markers that appeared most promising from published literature. We sequentially evaluated immunostaining with these markers in (1) surgical resection specimens, (2) cell blocks from EUS-FNA, and (3) direct smears of EUS-FN aspirates. Finally, we performed a combined evaluation of cytology and immunostaining in direct smears that were difficult to interpret and required a second consultative cytologic opinion.

Results: In resection specimens, the majority of pancreatic adenocarcinomas expressed all 5 markers but fascin, maspin, and carcinoembryonic antigen-related cell adhesion molecule 6 also were expressed abnormally in normal pancreata and in chronic pancreatitis. Further evaluation therefore was limited to the other 2 markers: mesothelin and 14-3-3sigma. It was feasible and useful to perform immunostaining in cell blocks and direct smear for diagnosing pancreatic cancer. In cases requiring a second cytologic consultation, a combined evaluation of cytologic morphology and immunostaining had 90% accuracy for a pancreatic adenocarcinoma diagnosis.

Conclusions: In conclusion, immunostaining with newer protein markers is feasible in EUS-FNA specimens and can assist cytopathologists in diagnosing pancreatic cancer. Among the currently available protein immunomarkers, a combination of mesothelin and 14-3-3sigma seems most promising, but needs to be validated in prospective studies before routine clinical use.