Endoscopic Ultrasound–Guided Fine Needle Aspirate Microsatellite Loss Analysis and Pancreatic Endocrine Tumor Outcome

Abstract

The clinical course of pancreatic endocrine tumor (PET) varies depending on tumor aggressiveness, disease extent, and possibly accumulated molecular alterations. Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) results, although accurate in diagnosing PET, correlate poorly with PET outcome. The role of detecting key molecular abnormalities in predicting PET behavior and clinical outcome from EUS-FNA material remains unknown. Patients with confirmed PET who underwent EUS-FNA during a 32-month period were included. Patient demographics and clinical data were recorded and follow-up information was obtained by contacting their physician to evaluate disease progression. Representative tumor cells were microdissected from the FNA material. DNA was harvested and amplified, targeting a panel of 17 polymorphic microsatellite markers on chromosomes 1p, 3p, 5q, 9p, 10q, 11q, 17p, 17q, 21q, and 22q. The polymerase chain reaction products were subjected to fluorescent capillary gel electrophoresis to detect microsatellite loss. The fractional allelic loss (FAL) was calculated. Twenty-five patients were studied. Thirteen were classified histologically as benign PET limited to the pancreas and 12 as malignant PET (invasive or metastatic). The mean FAL in the benign and malignant PET was 0.03 and 0.37 (P<.0001), respectively. In addition, the mean FAL was significantly greater in those with disease progression as compared with patients with stable disease (0.45 vs 0.09 respectively, P<.0001). Microsatellite loss analysis of EUS-FNA material from PET can be performed reliably and an FAL value of more than 0.2 is associated with disease progression. This technique may have value in the preoperative assessment and risk stratification of patients with PET.