OBJECTIVE: To investigate the expression of mesothelin in specimen of pancreas fine-needle aspiration and to evaluate the potential contribution of immunohistochemical labeling of mesothelin to the interpretation of pancreas fine-needle aspiration (FNA). METHODS: Specimens from 27 patients were selected for immunolabeling. Immunohistochemical EnVision method was used to detect the expression of mesothelin in specimen of pancreas fine-needle aspiration. The labeling in each patient was scored as positive or negative. These results were compared with the cytologic diagnosis and the follow-up data. RESULTS: Nineteen of the 27 patients were ultimately shown to have an adenocarcinoma, and 8 had no evidence of malignancy on follow-up. Initial cytologic diagnosis of malignancy correlated with carcinoma on follow-up in 10 of 10 cases, and initial benign cytologic diagnosis correlated with benign follow-up in 4 of 6 cases. Seven of the 11 patients with suspicious cytology were found to have carcinomas on follow-up. Mesothelin labeling was seen in 14 of the 19 patients ultimately shown to have carcinomas and was absent in 7 of the 8 benign lesions (sensitivity, 73.7%; specificity, 87.5%). Five of the 7 cytologically suspicious cases with malignant follow-up labeled for mesothelin. Positive mesothelin labeling was observed in one of the 4 suspicious cases who finally proved to be benign during follow-up. CONCLUSION: Immunohistochemical labeling for mesothelin may be a highly specific tool for the detection of pancreatic adenocarcinoma in FNA specimens and is useful in categorizing cytologically suspicious lesions.
OBJECTIVE: To investigate the expression of mesothelin in specimen of pancreas fine-needle aspiration and to evaluate the potential contribution of immunohistochemical labeling of mesothelin to the interpretation of pancreas fine-needle aspiration (FNA). METHODS: Specimens from 27 patients were selected for immunolabeling. Immunohistochemical EnVision method was used to detect the expression of mesothelin in specimen of pancreas fine-needle aspiration. The labeling in each patient was scored as positive or negative. These results were compared with the cytologic diagnosis and the follow-up data. RESULTS: Nineteen of the 27 patients were ultimately shown to have an adenocarcinoma, and 8 had no evidence of malignancy on follow-up. Initial cytologic diagnosis of malignancy correlated with carcinoma on follow-up in 10 of 10 cases, and initial benign cytologic diagnosis correlated with benign follow-up in 4 of 6 cases. Seven of the 11 patients with suspicious cytology were found to have carcinomas on follow-up. Mesothelin labeling was seen in 14 of the 19 patients ultimately shown to have carcinomas and was absent in 7 of the 8 benign lesions (sensitivity, 73.7%; specificity, 87.5%). Five of the 7 cytologically suspicious cases with malignant follow-up labeled for mesothelin. Positive mesothelin labeling was observed in one of the 4 suspicious cases who finally proved to be benign during follow-up. CONCLUSION: Immunohistochemical labeling for mesothelin may be a highly specific tool for the detection of pancreatic adenocarcinoma in FNA specimens and is useful in categorizing cytologically suspicious lesions.