PURPOSE: Mutations involved in neoplastic progression may be able to serve as markers for the presence of small numbers of neoplastic cells that would otherwise escape detection in diagnostic assays. Previous retrospective studies have suggested that the sensitivity of the cytologic diagnosis of pancreatic and biliary tract carcinomas is improved when analysis includes Ki-ras exon 1, which is commonly mutated in these neoplasms. We report our experience with the systematic prospective application of Ki-ras gene analysis to the evaluation of fine-needle aspirates and brushings from the pancreatobiliary tract. MATERIALS AND METHODS: Between September 1996 and April 1999, 75 pancreatic fine-needle aspirates and common bile duct brushings submitted for routine cytologic diagnosis were also evaluated for mutations in Ki-ras exon 1 by polymerase chain reaction/single-strand conformation polymorphism analysis. After routine preparation of the specimens, residual material was used for molecular analysis. Results are compared with the morphologic diagnosis and available clinical information. RESULTS: Single-strand conformation polymorphism mutation patterns in Ki-ras were detected in 22 of the 70 consecutive clinical specimens with adequate DNA and at least 6 months of available clinical follow-up. Sensitivity, specificity, and positive predictive value for the presence of concurrent or subsequent pancreatobiliary carcinoma were 33%, 97%, and 93%, respectively, for definitive cytologic diagnosis alone, and 53%, 97%, and 95% for positive Ki-ras single-strand conformation polymorphism mutation pattern alone. If definitive positive cytology or atypical/suspicious cytology with a positive Ki-ras single-strand conformation polymorphism mutation pattern is used, sensitivity is 55%, specificity is 97%, and positive predictive value is 96% for the presence of pancreatobiliary carcinoma. DISCUSSION: Results support the routine use of Ki-ras mutational analysis to increase the sensitivity of the cytologic evaluation of pancreatobiliary fine-needle aspirates and common bile duct brushings with atypical or suspicious morphology without compromising specificity.
PURPOSE: Mutations involved in neoplastic progression may be able to serve as markers for the presence of small numbers of neoplastic cells that would otherwise escape detection in diagnostic assays. Previous retrospective studies have suggested that the sensitivity of the cytologic diagnosis of pancreatic and biliary tract carcinomas is improved when analysis includes Ki-ras exon 1, which is commonly mutated in these neoplasms. We report our experience with the systematic prospective application of Ki-ras gene analysis to the evaluation of fine-needle aspirates and brushings from the pancreatobiliary tract. MATERIALS AND METHODS: Between September 1996 and April 1999, 75 pancreatic fine-needle aspirates and common bile duct brushings submitted for routine cytologic diagnosis were also evaluated for mutations in Ki-ras exon 1 by polymerase chain reaction/single-strand conformation polymorphism analysis. After routine preparation of the specimens, residual material was used for molecular analysis. Results are compared with the morphologic diagnosis and available clinical information. RESULTS: Single-strand conformation polymorphism mutation patterns in Ki-ras were detected in 22 of the 70 consecutive clinical specimens with adequate DNA and at least 6 months of available clinical follow-up. Sensitivity, specificity, and positive predictive value for the presence of concurrent or subsequent pancreatobiliary carcinoma were 33%, 97%, and 93%, respectively, for definitive cytologic diagnosis alone, and 53%, 97%, and 95% for positive Ki-ras single-strand conformation polymorphism mutation pattern alone. If definitive positive cytology or atypical/suspicious cytology with a positive Ki-ras single-strand conformation polymorphism mutation pattern is used, sensitivity is 55%, specificity is 97%, and positive predictive value is 96% for the presence of pancreatobiliary carcinoma. DISCUSSION: Results support the routine use of Ki-ras mutational analysis to increase the sensitivity of the cytologic evaluation of pancreatobiliary fine-needle aspirates and common bile duct brushings with atypical or suspicious morphology without compromising specificity.