ABSTRACT
【Objective】 To investigate the potential genes and pathways associated with esophageal adenocarcinoma through microarray expression profiling data analysis and bioinformatics approaches. 【Methods】 The mRNA expression microarray data related to esophageal adenocarcinoma development were screened out with GEO database, and the biological processes, signaling pathways and network of these genes were statistically analyzed using "R" software. 【Results】 The GSE26886 was obtained from GEO database. A total of 1383 differentially expressed genes were associated with carcinogenesis of esophageal adenocarcinoma, including 607 up-regulated and 776 down-regulated genes. These genes were involved in metabolism, stimulate responses, cell adhesion, cell regeneration and immune biological processes. Eight significantly enriched pathways were identified by pathway analysis. 【Conclusion】 The bioinformatic method can analyze the gene chip data effectively. Multiple genes and signaling pathways are involved in the carcinogenesis of esophageal adenocarcinoma, which provides a new idea or approach for exploring biomarkers of early screening and therapeutic targets.
ABSTRACT
Objective: To explore the risk of incidence of catheter-related thrombosis in cancer patients receiving chemotherapy using nomogran. Methods: We retrospectively evaluated 286 patients with malignant tumor who needed PICC insertion at the First Affiliated Hospital of Xi'an Jiaotong University between December 2014 and December 2015. Potential risk factors were included in the least absolute shrinkage and selection operator (LASSO) regression analysis to finally build a nomogram to predict the risk of PICC-related thrombosis. Results: A total of 286 patients who needed PICC insertion were analyzed, among whom 72 experienced PICC- related thrombosis. Twenty-seven potential thrombosis-related risk factors were included in the LASSO regression analysis. The results indicated that the use of ultrasound guidance during insertion, previous chemotherapy, other catheter-related complications, and plasma Ddimer were the risk factors of PICC related thrombosis. Thus, these four risk factors were applied to the nomogram model. Further,the nomogram prediction model yielded a C-index of 0.688 and the adjusted fitting curve was located in the error range of 10%. Conclusions:Combined with puncture technology, previous chemotherapy history, complications, and D-dimer level constituted the nomogram prediction model for PICC-related thrombosis which had a good accuracy.