CIBERSORT反卷积算法计算胃腺癌免疫浸润的研究

Calculation of immune infiltration of gastric adenocarcinoma by CIBERSORT deconvolution algorithme

目的对胃腺癌的免疫浸润细胞进行分析,以期寻找胃腺癌的免疫治疗靶点。方法在公共数据库TCGA中进行检索,采用Perl脚本对数据进行整理,共纳入348例患者,包含343份肿瘤样本和30份正常样本,采用CIBERSORT反卷积算法估计肿瘤组织中不同免疫细胞占比,最后利用R软件进行差异免疫细胞分析、生存分析、临床相关性分析。结果肿瘤样本中未活化CD4记忆T细胞、CD8 T细胞、M2巨噬细胞和M0巨噬细胞的免疫浸润最明显,其占比分别为15.90%、12.67%、11.81%和12.05%。而差异分析表明浆细胞、单核细胞、活化CD4记忆T细胞、M0巨噬细胞、M1巨噬细胞和M2巨噬细胞占比在正常样本和肿瘤样本中差异有统计学意义(P<0.05)。生存分析发现,免疫浸润中低表达的未活化CD4记忆T细胞可有效提高生存率(P=0.029),而高表达的CD8 T细胞可有效提高生存率(P=0.002)。从免疫细胞的临床相关性发现,未活化树突细胞和CD8 T细胞随着胃腺癌分级的增加,浸润会逐渐增加,差异有统计学意义(P<0.05)。而T细胞、M0巨噬细胞、活化肥大细胞和中性粒细胞随着分级的增加,浸润会逐渐减轻,差异有统计学意义(P<0.05)。结论利用CIBERSORT方法可有效地筛选出胃腺癌免疫细胞占比,未活化CD4记忆T细胞、CD8 T细胞、M2巨噬细胞和M0巨噬细胞在胃腺癌肿瘤浸润和预后过程中发挥了重要作用,有望将免疫浸润较多的免疫细胞作为药物靶标来提高患者的生存率。

Objective To analyze immune infiltrating cells of gastric adenocarcinoma in order to find immunotherapy targets for gastric adenocarcinoma.Methods The search was performed in the public database TCGA,and the data was sorted using Perl script.This study included 348 patients,including 343 cancer samples and 30 normal samples.The CIBERSORT deconvolution algorithm was used to estimate the different immune cells in the tumor tissue.In the end,the R software was used for differential immune cell analysis,survival analysis,and clinical correlation analysis.Results In the tumor samples,the immune infiltration of the inactive CD4 memory T cells,CD8 T cells,M2 macrophages and M0 macrophages was the most obvious,accounting for 15.90%,12.67%,11.81%and 12.05%.The difference analysis showed that the proportion of plasma cells,monocytes,activated CD4 memory T cells,M0 macrophages,M1 macrophages and M2 macrophages in normal and tumor samples were significantly different(P<0.05).Survival analysis showed that the low expression of inactive CD4 memory T cells in immune infiltration could effectively improve the survival rate(P=0.029),while the high expression of CD8 T cells could effectively improve the survival rate(P=0.002).From the clinical correlation of immune cells,it was found that the infiltration fraction of unactivated dendritic cells and CD8 T cells increased with the increase of gastric adenocarcinoma grade(P<0.05).With the increase of T cells,M0 macrophages,active mast cells and neutrophils,the infiltration fraction would gradually decrease(P<0.05).Conclusion The CIBERSORT method can effectively screen out the proportion of immune cells in gastric adenocarcinoma.Unactivated CD4 memory T cells,CD8 T cells,macrophages M2 and macrophages M0 play an important role in tumor invasion and prognosis of gastric adenocarcinoma.It is expected that immune cells with more immune infiltration will be used as drug targets to improve the survival rate of patients.