Next-generation sequencing mutation analysis on biliary brush cytology for differentiation of benign and malignant strictures in primary sclerosing cholangitis

• Published in: Gastrointestinal endoscopy Vol. 97; no. 3; pp. 456 - 465.e6
• Main Authors: Kamp, Eline J.C. A., Dinjens, Winand N.M., van Velthuysen, Marie-Louise F., de Jonge, Pieter Jan F., Bruno, Marco J., Peppelenbosch, Maikel P., de Vries, Annemarie C.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.03.2023
• Subjects: Bile Duct Neoplasms - pathology; Bile Ducts, Intrahepatic; Biliary Tract - pathology; Case-Control Studies; Cell Differentiation; Cholangiocarcinoma - complications; Cholangiocarcinoma - genetics; Cholangitis, Sclerosing - complications; Cholangitis, Sclerosing - genetics; Cholestasis - pathology; Constriction, Pathologic - pathology; Cytodiagnosis; High-Throughput Nucleotide Sequencing; Humans; CCA; LOD; FISH; PSC; NGS; PSC-CCA
• ISSN: 0016-5107; 1097-6779; 1097-6779
• DOI: 10.1016/j.gie.2022.10.014

Differentiation of benign and malignant biliary tract strictures on brush material remains highly challenging but is essential for adequate clinical management of patients with primary sclerosing cholangitis (PSC). In this case-control study, biliary brush cytology samples from PSC patients with cholangiocarcinoma (PSC-CCA) were compared with samples from PSC patients without CCA (PSC-control subjects) using next-generation sequencing (NGS). Cells on archived slides were dissected for DNA extraction. NGS was performed using a gene panel containing 242 hotspots in 14 genes. Repeated brush samples from the same patient were analyzed to study the consistency of NGS results. In PSC-CCA cases that underwent surgical resection, molecular aberrations in brush samples were compared with NGS data from subsequent resection specimens. Forty patients (20 PSC-CCA and 20 PSC-control subjects) were included. The gene panel detected 22 mutations in 15 of 20 PSC-CCA brush samples, including mutations in TP53 (8 brush samples), K-ras (5), G-nas (3), ERBB2 (1), APC (1), PIK3CA (1), and SMAD4 (1). One G-nas and 3 K-ras mutations were found in 3 of 20 PSC-control brush samples. The sensitivity of the NGS panel was 75% (95% confidence interval, 62%-80%) and specificity 85% (95% confidence interval, 64%-95%). Repeated brush samples showed identical mutations in 6 of 9 cases. Three repeated brush samples demonstrated additional mutations as compared with the first brush sample. In 6 of 7 patients, mutations in brush samples were identical to mutations in subsequent resection specimens. NGS mutation analysis of PSC brush cytology detects oncogenic mutations with high sensitivity and specificity and seems to constitute a valuable adjunct to cytologic assessment of brush samples. [Display omitted]