ABSTRACT
Objective To study the molecular genetic characteristics and immunophenotype of diffuse large B -cell lymphoma (DLBCL),thus to provide evidence in further research.Methods 92 patients with DLBCL were selected.Immunophenotype was classified by Hans method,bcl -2 and bcl -6 were detected by fluorescence in situ hybridization(FISH).The relationship between immunophenotype and genes expression was analyzed by SPSS 19.0. Results There were GCB with 28 cases(30.43%)and non GCB with 64 cases(69.57%).Bcl -2 gene abnormality was more found in GCB(χ2 =38.39,P 0.05).There was no correlation between bcl -2 translocation and bcl -6 translocation(r =0.095,P >0.05),but there was significant correlation between bcl -2 amplification and bcl -6 translocation(r =0.750,P 0.05 ).Conclusion Bcl -2 gene abnormality was more found in GCB,but the relationship between bcl -2 and bcl -6 is not clear.There may be no correlation between location and molecular genetic.
ABSTRACT
Objective To study the clinical, pathological and immunohistochemical characteristics of gastrointestinal stromal tumors (GISTs). Methods 87 cases of GISTs were studied with conventional histopathological and immunohistochemical stains. Results All patients were adults, age range 28 to 73 years (mean 53 year), the tumors were situated in stomach (56 cases), in small intestine (29 cases), in colon (2 cases). The tumors had two kinds of cells: spindle cells and epithelioid cells, they were different in proportion. Immunohistochemically, vimentin were all positive, CD117 and CD34 showed diffuse strong expression, SMA and S-100 were positive when the tumors differentiated to smooth muscle and nerve. Conclusion GISTs predominantly occurred in middle aged or old patients, the tumors were composed of spindle cells and epithelioid cells. The immunohistochemical characters were positive for CD117 and CD34.
ABSTRACT
The relationship between the precision of the estimated sampling variances and the number of samples was quantitatively established. It revealed that the product of the standard deviation of the estimated samplingvariance and square root of the number of samples was a constant. Monte Carlo simulation technique was employed to verify the equation. Populations with Gaussian,unifohn random and multi-nomial(taking silicon carbide material as an example) distributions were studied with satisfactory results.