Insulin-like Growth Factor II mRNA-Binding Protein 3 Expression in Benign and Premalignant Lesions and Carcinomas of the Stomach

BACKGROUND: Recent studies have demonstrated that insulin-like growth factor II mRNA-binding protein 3 (IMP3) is expressed in malignant tumors of various organs but not in normal tissue. We investigated IMP3 expression in various benign lesions, premalignant lesions and carcinomas of the stomach. METHODS: IMP3 immunohistochemical staining was performed on 24 benign gastric lesions, 24 gastric adenomas, and 322 gastric carcinomas. RESULTS: IMP3 was not expressed in benign gastric lesions including adenomas with low-grade dysplasia, but was expressed in 17% of adenomas with high-grade dysplasia, and in 44% of carcinomas. As the carcinomas were in the advanced stage, they expressed IMP3 more frequently and strongly. Patients with IMP3-positive tumors had poorer survival than those with negative tumors. CONCLUSIONS: IMP3 expression in gastric carcinoma may be related to tumor invasion and metastasis, and is an independent risk factor for poor prognosis.

IMP3 Expression of the Cholangiocarcinoma in Cytology Specimen and Its Diagnostic Value

BACKGROUND: Pathologists occasionally have difficulty in distinguishing cholangiocarcinoma from benign reactive cells on the biliary cytology. We estimated the diagnostic value of insulinlike growth factor II mRNA binding protein 3 (IMP3) on biliary cytology specimens. METHODS: We selected 40 patients who had undergone a biliary cytology examination and whose diagnoses were confirmed by biopsy. The IMP3 expression was evaluated by immunohistochemical staining in both the tissue and the cytology specimens. Their histologic diagnoses were cholangiocarcinoma (n = 34), high grade dysplasia (n = 2), low grade dysplasia (n = 1) and non-biliary tumor or no tumor (n = 3). RESULTS: On the histology, all the cases of cholangiocarcinoma or high grade dysplasia were positive for IMP3 (36/36). The low grade dysplasia or the normal bile ducts in the other 4 cases were negative for IMP3 (0/4). Of the 36 cases diagnosed as cholangiocarcinoma or high grade dysplasia histologically, 29 cases were positive for IMP3 on the cytology (29/36; sensitivity, 80.6%). Of the 4 cases without cholangiocarcinoma or high grade dysplasia, 1 case was weakly positive for IMP3 on the cytology. The histologic diagnosis of this case was hepatocellular carcinoma. Except for this hepatocellular carcinoma case, the specificity of IMP3 on the cytology was 100%. CONCLUSIONS: IMP3 is a useful diagnostic marker for cholangiocarcinoma on cytology.