Objective To investigate expressions of discovered on GIST-1（DOG1），CD117 and insulin-like growth factor 1 receptor（IGF1R），and the mutation status of c-KIT and PDGFRA gene in gastrointestinal stromal tumors （GISTs），as well as the relations between them and clinicopathologic features of GISTs.Methods The expressions of DOG1，CD117 and IGF1R were detected by immunohistochemistry in 98 GISTs and 40 non-GISTs. Mutations of c-KIT and PDGFRA were analyzed using a matrixassisted laser desorption／ionizationtime of flight（MALDI-TOF）mass spectrometer.Results The positive expression rates of DOG1，CD117 and IGF1R in GISTs were 92.9％，95.9％ and 6.1％，respectively. And in non-GISTs，the positive rates of DOG1 and CD117 were 10.0％ and 17.5％，lower than those in GISTs（P＜0.001）. The expression of DOG1 was correlated with tumor size and risk degree. For CD117，its expression was related to tumor location and histological types. IGF1R was not significantly associated with clinicopathologic features. Of 77 GISTs，mutation rates of c-KIT and PDGFRA were 64.9％ and 22.1％，and the most common mutational sites were exon 11（54.0%）and 12（16.9%）respectively. The mutation rates of PDGFRA and c-KIT were only related to tumor location（P=0.001，P=0.002）. Expressions of CD117 and DOG1 were not related to mutations of c-KIT or PDGFRA gene. But IGF1R's expression was significantly different between wild-type and mutated GISTs.Conclusion Detection of both DOG1 and CD117 can raise the diagnostic rate of GISTs，especially for those who has negative expression of CD117 and the results can help analyze clinicopathologic features of GISTs comprehensively. Our results suggest that IGF1R is a special marker and potential the rapeutic target for wild-type GISTs. There will be more useful to analyze response to tyrosine kinase inhibitors and prognosis of GISTs with gene mutation analysis and tumor locations.

Objective To investigate the expression of hypoxiainducible factor-1α（HIF-1α） in nonsmall cell lung cancer（NSCLC）and its relationship with clinicopathological feature with a systematic review. Methods Studies assessing clinical significance of HIF-1α expression in NSCLC published until April 2012 were selected on the basis of Pubmed，EMBASE，Cochrane Library，and CNKI. The meta analysis in this study was performed by the statistical software STATA 11.0.A meta analysis was performed to clarify the impact of HIF-1α expression on clinicopathological in NSCLC. Results A total of 19 studies were recruited. As for the positive rate of HIF-1α expression，significant differences were tested between NSCLC and normal lung tissues（OR=35.55，95％CI: 18.79-67.25，P=0.000）. There was no significant differences between T stages T3-T4 and T stages T1-T2（OR=2.50，95％CI:0.89-6.97，P=0.081）. There were significant differences between low differentiation and medium-and high-differentiation（OR=1.81，95％CI:1.11-2.98，P＜0.05），lymph node metastasis and non-lymph node metastasis（OR=2.97，95％CI:1.84-4.80，P＜0.05），and clinic stages Ⅲ-Ⅳ and clinic stagesⅠ-Ⅱ（OR=2.56，95％CI:1.58-4.15，P＜0.05）. Conclusion HIF-1α may play an important role in NSCLC，and the overexpression of HIF-1α correlates with malignant biological behavior.