Accuracy of Endoscopic Ultrasonographic Impression Compared with Pathologic Diagnosis in Gastrointestinal Submucosal Tumors / 대한소화기학회지

BACKGROUND/AIMS: Endoscopic ultrasonography (EUS) is a valuable imaging modality for the evaluation of gastrointestinal submucosal tumor (SMT). EUS is helpful in assessing the layer of origin, tumor diameter, shape, border characteristics, and internal echo patterns of SMTs and thus makes it possible to predict histologic diagnosis with educated guess. However, some studies have found no significant differences in EUS features between benign and malignant mesenchymal tumors. By comparing EUS impressions with histologic diagnosis, we evaluated the accuracy of EUS in differential diagnosis of gastrointestinal SMTs. METHODS: 58 cases of gastrointestinal SMTs with both EUS findings and pathologic reports were compared retrospectively from August 2001 to September 2003. RESULTS: 34 patients had lesions in the stomach and 13, 8, 3 in the esophagus, duodenum, and colon respectively. Benign lesions were predominant (46 of 58). The EUS and pathologic diagnosis coincided in 46/58 (79.3%) of the cases. Use of EUS led to the correct diagnosis in 7/9 (77.8%) of malignant GISTs (gastrointestinal stromal tumor) and leiomyosarcomas. Two small malignant gastric GISTs were diagnosed as benign with EUS. CONCLUSIONS: EUS is a useful tool in the differential diagnosis of gastrointestinal SMTs and predicting malignant lesions. However, some malignant GISTs were diagnosed as benign tumor with EUS examination.

Effects of TNF-alpha on Angiogenesis in CT-26 Mouse Colon Carcinoma Homograft Model / 대한소화기학회지

BACKGROUND/AIMS: Tumor necrosis factor-alpha (TNF-alpha) exerts its anti-tumor effect through direct cytotoxicity on tumor cells and damage to the tumor vasculature. However, its role in tumor angiogenesis is controversial. We evaluated the angiogenic effect of TNF-alpha on BALB/c mouse colon carcinoma homograft model. METHODS: Ten BALB/c mice were inoculated intraperitoneally with CT-26 mouse colon carcinoma cells. After a week, recombinant mouse TNF-alpha (2microgram/mL) were given four times on every other day to five animals and the same volume of phosphate buffered saline was given at the same interval to five animals as control. Harvested tumor tissues were stained by immunohistochemistry with CD31 and VEGF antibodies. Number of microvessels and VEGF expression were counted by light microscope. RESULTS: The mean microvessel counts per 200x field of TNF-alpha treated animals were 70.2+/-7.8 and those of nontreated animals were 83.8+/-8.3 (p<0.05). The VEGF score of both groups were 3. CONCLUSIONS: TNF-alpha treated animals showed decreased microvessel counts in tumor tissue but VEGF expression in both groups showed no difference. Therefore, TNF-alpha showed antiangiogenic effects on colon carcinoma homograft model.