BACKGROUND & AIMS: Mutations at codon 12 of the K-ras gene are present in 65%-100% of carcinomas of human exocrine pancreas and could be used as a potential tumor marker at the tissue level. The purpose of this study was to assess, in large series of patients, the utility of K-ras mutation analysis to evaluate fine-needle aspirates of pancreatic masses. METHODS: One hundred fifteen fine-needle aspirates obtained from 93 patients were evaluated retrospectively. Cytological analysis was based on the review of cell blocks. Mutations were detected by using artificial restriction fragment length polymorphisms using the Hphl and BstNl approaches. RESULTS: The sensitivity and specificity of cell block cytology was 64% and 100%, respectively, for the diagnosis of pancreatic carcinoma. K-ras mutations were detected in 41 pancreatic carcinomas (sensitivity, 59%) and in one mucinous cystic tumor; specificity of ras analysis alone was 100%. The sensitivity of cytology combined with K-ras mutations were 77.6% and 100%, respectively. CONCLUSIONS: The detection of K-ras mutations would have suggested the diagnosis of pancreatic cancer in 14 cases otherwise not detected by cytology alone. K-ras mutation analysis should be restricted to cell blocks containing suspicious, normal-appearing duct cells, or insufficient material in the cytological examination.
BACKGROUND & AIMS: Mutations at codon 12 of the K-ras gene are present in 65%-100% of carcinomas of human exocrine pancreas and could be used as a potential tumor marker at the tissue level. The purpose of this study was to assess, in large series of patients, the utility of K-ras mutation analysis to evaluate fine-needle aspirates of pancreatic masses. METHODS: One hundred fifteen fine-needle aspirates obtained from 93 patients were evaluated retrospectively. Cytological analysis was based on the review of cell blocks. Mutations were detected by using artificial restriction fragment length polymorphisms using the Hphl and BstNl approaches. RESULTS: The sensitivity and specificity of cell block cytology was 64% and 100%, respectively, for the diagnosis of pancreatic carcinoma. K-ras mutations were detected in 41 pancreatic carcinomas (sensitivity, 59%) and in one mucinous cystic tumor; specificity of ras analysis alone was 100%. The sensitivity of cytology combined with K-ras mutations were 77.6% and 100%, respectively. CONCLUSIONS: The detection of K-ras mutations would have suggested the diagnosis of pancreatic cancer in 14 cases otherwise not detected by cytology alone. K-ras mutation analysis should be restricted to cell blocks containing suspicious, normal-appearing duct cells, or insufficient material in the cytological examination.