基于生物信息学分析POSTN 基因在胃癌中的表达及免疫意义.

Objective: To investigate value of periostin （POSTN） in immunity and prognosis of gastric cancer. Methods: POSTN expression in gastric cancer and its relationship with survival and prognosis were analyzed by bioinformatics methods. Distribution of immune cells in cancer and normal tissues in TCGA gastric cancer cohort were compared. POSTN-related immunomodulators were extracted from TISIDB database, and POSTN-related immune events were analyzed by GO and KEGG pathway enrichments. Based on POSTN associated immunomodulators, a polygenic risk prediction model was developed by Cox regression, from which risk scores were generated. Prediction accuracy of the model was evaluated by subject operating characteristic curve. Prognostic normograms with calibration curves were constructed to predict 5-year survival probability of individuals. Results: POSTN expression in gastric cancer tissues was significantly higher than that in non-tumor tissues. Overall survival of POSTN high expression group was shorter than that of POSTN low expression group. High expression of POSTN was positively correlated with high level of immune infiltration in most immune cells. KEGG pathway analysis based on 46 POSTN-associated immunomodulators revealed involvement of NF-κB signaling and cell adhesion molecules. Risk score could be used as an independent predictor of prognosis in patients with gastric cancer. Conclusion: POSTN is highly expressed in gastric cancer tissues, whose expression is correlated with overall survival of gastric cancer patients. POSTN is associated with tumor immune invasion, suggesting that POSTN may be a potential immunotherapy target for gastric cancer. [ABSTRACT FROM AUTHOR]