摘要目的:构建缺氧和免疫相关风险评分模型来预测胃癌(gastric cancer, GC)患者的预后。方法:通过无监督共识聚类算法生成两个不同的低氧聚类和免疫聚类，基于哈尔滨医科大学-GC队列使用LIMMA包分析筛选出2227个缺氧相关和2640个免疫相关的胃癌差异表达基因。单变量Cox回归生存分析结合LASSO-Cox回归确定5个缺氧和免疫相关特征预后基因(FAIM2、CMA1 、MMP12、C11orf53、FBLN5)并构建风险评分模型。结果:构建的缺氧和免疫相关预后特征基因是GC预后的独立因素( P<0.001)，并进一步构建预后模型。通过基因富集分析这些特征基因的生物学功能。通过CIBERSORT算法分析免疫差异。 结论:确定了一种针对缺氧和免疫相关预后特征可用作对GC风险进行分层的方法。

abstractsObjective:To build a model of hypoxia- and immune-related risk score to predict the prognosis of gastric cance(rGC) patients.Methods:Two different hypoxia clusters and immune clusters were generated through unsupervised consensus clustering algorithms, and 2227 hypoxic-related and 2640 immune-related gastric cancer differentially expressed genes were screened by LIMMA package analysis based on the Harbin Medical University-GC cohort. Univariate Cox regression survival analysis combined with LASSO-Cox regression was used to identify five prognostic genes (FAIM2, CMA1, MMP12, C11orf53, FBLN5) of hypoxia and immune-related characteristics and to construct a risk scoring model.Results:The anoxic and immune-related prognostic characteristic genes constructed in this study are independent factors of GC prognosis ( P<0.001), and further construct the prognosis model. The biological function of these characteristic genes was analyzed by gene enrichment. The immune infiltration difference was analyzed by CIBERSORT algorithm. Conclusions:The method that can be used to stratify the risk of GC according to the prognosis characteristics related to hypoxia and immunity is identified.