Fluorescent in situ Hybridization in the Diagnosis of Indeterminate Biliary Strictures

Abstract

Purpose: Cholangiocarcinoma (CCA) diagnosis has been limited by poor sensitivity of biliary brush cytology. Fluorescent in situ hybridization allows early detection of chromosomal aneuploidy or deletions in biliary cells and has the potential to improve diagnostic sensitivity compared to standard cytology. Methods: All patients referred for ERCP of indeterminate biliary strictures who underwent biliary brushing between 2/2008 and 4/2009 at our institution were included. We limited the analysis of test performance to those cases where final pathology was available or where follow-up data after a 6 month interval was obtained. A cytology result was considered positive if read as “suspicious” or “malignant” by a cytopathologist. FISH was performed using CEP 3, CEP 7, and CEP 17 probes for chromosome enumeration and 9p21 (p16) for deletion. FISH result was considered positive if aneuploidy or deletion was detected with these probes. Statistical comparison between the techniques was performed with area-under-ROC-curve (AUC) analysis. Results: 26 patients underwent ERCP, cytology, and FISH during this period. 18 underwent surgery or reached a final diagnosis and 6 month follow-up was available for 8. 12 cholangiocarcinomas, 2 pancreatic adenocarcinomas, 3 primary sclerosing cholangitis, and 1 benign stricture were diagnosed. Cytology had a sensitivity and specificity of 21% and 100%, respectively, for diagnosing malignancy. FISH had a sensitivity and specificity of 86% and 100%, respectively. Cytology and FISH (considered positive when either were positive) combined to yield a sensitivity of 93%, specificity of 100%, positive predictive value (PPV) of 100% and negative predictive value (NPV) of 92%. The AUC for the combination of cytology and FISH was significantly better than for cytology alone (p<0.001). Conclusion: FISH, when added to biliary brush cytology, significantly improved the diagnostic accuracy of detecting malignancy in indeterminate strictures. This study further validates previous findings that FISH should be employed as a standard analysis for all biliary brush specimens in patients with indeterminate biliary strictures.Table