Next-Generation Sequencing and Fluorescence in Situ Hybridization Have Comparable Performance Characteristics in the Analysis of Pancreaticobiliary Brushings for Malignancy

Abstract

Cytological evaluation of pancreatic or biliary duct brushings is a specific, but insensitive, test for malignancy. We compared adjunctive molecular testing with next-generation sequencing (NGS) relative to fluorescence in situ hybridization (FISH) for detection of high-risk neoplasia or malignancy. Bile duct brushings from 81 specimens were subjected to cytological analysis, FISH using the UroVysion probe set, and targeted NGS. Specimens were placed into negative/atypical (negative) or suspicious/positive (positive) categories depending on cytology and negative or positive categories on the basis of FISH and NGS results. Performance characteristics for each diagnostic modality were calculated on the basis of clinicopathologic follow-up and compared in a receiver operating characteristic analysis. There were 33 high-risk neoplasia/malignant strictures (41%) and 48 benign (59%). NGS revealed driver mutations in 24 cases (30%), including KRAS (21 of 24 cases), TP53 (14 of 24 cases), SMAD4 (6 of 24 cases), and CDKN2A (4 of 24 cases). Cytology had a sensitivity of 67% (95% CI, 48%-82%) and a specificity of 98% (95% CI, 89%-100%). When added to cytology, NGS increased the sensitivity to 85% (95% CI, 68%-95%), leading to a significant increase in the area under the curve in a receiver operating characteristic analysis (P = 0.03). FISH increased the sensitivity to 76% (95% CI, 58%-89%), without significantly increasing the area under the curve. These results suggest that ancillary NGS testing offers advantages over FISH, although studies with larger cohorts are needed to verify these findings.