目的 基于生物信息学方法分析富含丝氨酸/精氨酸的剪接因子9（SRSF9）在头颈部鳞状细胞癌（HNSCC）中的表达情况、临床意义以及与肿瘤免疫浸润的关系，并探讨其作用机制。 方法 利用癌症基因图谱（TCGA）数据库、基因表达综合（GEO）数据库中GSE30784和GSE13601数据集、临床蛋白质组学肿瘤分析联盟（CPTAC）数据库、基因表达交互分析（GEPIA）数据库和Kaplan-Meier plotter数据库分析SRSF9在HNSCC中的表达及与患者临床病理特征和预后的关系；利用cBioPortal数据库分析SRSF9基因变异情况；应用ESTIMATE算法和肿瘤免疫估计资源（TIMER）数据库评估SRSF9表达与肿瘤微环境和肿瘤免疫细胞浸润的关系；利用LinkedOmics数据库分析SRSF9在HNSCC中的共表达基因及其调控通路；基于TCGA SpliceSeq数据库分析HNSCC中SRSF9调控的可变剪接事件，并对靶基因进行基因本体（GO）功能富集分析和京都基因与基因组百科全书（KEGG）信号通路富集分析。 结果 TCGA数据库、GEO数据库中GSE30784和GSE13601数据集分析，HNSCC组织中SRSF9 mRNA表达水平较癌旁正常组织明显升高（P<0.01）；CPTAC数据库分析，与正常组织比较，HNSCC组织中SRSF9蛋白表达水平明显升高（P<0.001）。TCGA数据库分析，SRSF9 mRNA表达与HNSCC患者病理分级（P=0.004）和HPV感染（P=0.031）有关联。GEPIA和Kaplan-Meier plotter数据库分析，HNSCC组织中SRSF9 mRNA高表达均与患者总生存期（OS）较差相关［风险比（HR）=1.40，P=0.019；HR=1.55，P=0.003］。cBioPortal数据库分析，HNSCC中SRSF9基因拷贝数变异（CNV）占26.85%，其CNV与SRSF9 mRNA表达水平呈正相关关系（r=0.44，P<0.001）。ESTIMATE算法分析，SRSF9 mRNA高表达组基质评分和免疫评分均较SRSF9 mRNA低表达组低（P<0.001），肿瘤纯度则高于SRSF9 mRNA低表达组（P<0.001）。TIMER数据库分析，SRSF9 mRNA表达水平与CD4+T淋巴细胞浸润呈正相关关系（r=0.186，P<0.001），而与B淋巴细胞、CD8+T淋巴细胞和树突状细胞浸润呈负相关关系（r=－0.269，P<0.001；r=－0.353，P<0.001；r=－0.304，P<0.001）。SRSF9共表达基因富集分析，核糖体、剪接体和代谢通路等相关基因上调，而黏着斑、细胞因子和细胞黏附分子等相关基因下调。SRSF9相关可变剪接靶基因主要涉及脂类代谢、胰高血糖素和紧密连接等通路。 结论 SRSF9在HNSCC中呈高表达，且与患者预后不良和肿瘤微环境免疫细胞浸润相关，可成为HNSCC诊断、预后评估和治疗的潜在分子靶标。

Objective To analyze the expression， clinical significance， and relationship with tumor immune infiltration of serine/arginine-rich splicing factor 9 （SRSF9） in head and neck squamous cell carcinoma （HNSCC） by bioinformatics methods， and to discuss its mechanism. Methods The expression of SRSF9 in HNSCC and its relationship with clinical pathologic characteristics and prognosis of the patients were analyzed by The Cancer Genome Atlas （TCGA） Database， GSE30784 and GSE13601 datasets in Gene Expression Omnibus （GEO） Database， Clinical Proteomic Tumor Analysis Consortium （CPTAC） Database， Gene Expression Profiling Interactive Analysis （GEPIA） Database， and Kaplan-Meier plotter Database；the variations of SRSF9 gene were examined through cBioPortal Database； ESTIMATE algorithm and The Tumor Immune Estimation Resource （TIMER） Database were used to assess the correlation between SRSF9 expression and tumor microenvironment， as well as tumor immune cell infiltration； LinkedOmics Database was used to analyze the SRSF9 co-expressed genes and their regulatory pathways in the HNSCC；the TCGA SpliceSeq Database was used to analyze the variable splicing events regulated by SRSF9 in the HNSCC；Gene Ontology （GO） functional enrichment analysis and Kyoto Encyclopedia of Genes and Genomes （KEGG） signaling pathway enrichment analysis were conducted on the target genes. Results The analysis results from TCGA Database and GSE30784 and GSE13601 datasets in GEO Database showed that the expression level of SRSF9 mRNA in the HNSCC tissue was significantly higher than that in adjacent normal tissue （P<0.01）.The CPTAC Database analysis results showed that compared with normal tissue， the expression level of SRSF9 protein in the HNSCC tissue was significantly increased （P<0.001）.The SRSF9 mRNA expression was associated with pathological grading （P=0.004） and HPV infection （P=0.031） in the patients with HNSCC according to TCGA Database analysis. The GEPIA and Kaplan-Meier plotter Database analysis results showed that high expression of SRSF9 mRNA in the HNSCC tissue was correlated with poorer overall survival （OS） of the patients ［hazard ratio （HR） = 1.40， P=0.019； HR=1.55， P=0.003］. The cBioPortal Database analysis results showed that copy number variation （CNV） of SRSF9 gene occurred in 26.85% in HNSCC， and CNV was positively correlated with SRSF9 mRNA expression levels （r=0.44， P<0.001）. The ESTIMATE algorithm analysis results showed that high expression of SRSF9 mRNA group had lower stromal and immune score， and higher tumor purity than those in low expression of SRSF9 mRNA group（P<0.001）. The TIMER Database analysis results showed there was a positive correlation between the expression of SRSF9 mRNA and CD4+ T lymphocyte infiltration （r=0.186， P<0.001）， and there were negative correlations between the expression level SRSF9 mRNA and B lymphocyte， CD8+ T lymphocyte， and dendritic cell infiltrations （r=－0.269， P<0.001； r=－0.353， P<0.001； r=－0.304， P<0.001）. The co-expressed gene enrichment analysis results showed upregulation of genes related to ribosomes， spliceosomes， and metabolic pathways， and downregulation of genes associated with focal adhesions， cytokines， and cell adhesion molecules. The main pathways involved in SRSF9-related variable splicing target genes were lipid metabolism， glucagon， and tight junctions. Conclusion SRSF9 is highly expressed in HNSCC and is associated with poor prognosis and tumor microenvironment immune cell infiltration， suggesting its potential as a molecular target for the diagnosis， prognosis assessment， and treatment of HNSCC.