基于生物信息学分析肿瘤浸润免疫细胞与头颈部鳞癌预后的关系

Bioinformatics-Based Analysis of the Relationship between Tumor-Infiltrating Immune Cells and the Prognosis of Head and Neck Squamous Cell Carcinoma

摘要: 目的：探究肿瘤浸润免疫细胞(TIICs)浸润程度与头颈部鳞状细胞癌(HNSCC)患者预后的关系，并基于多种免疫细胞构建预后模型。方法：从TCGA数据库中下载515例HNSCC样本和44例正常对照样本的基因表达谱和生存信息，利用CIBERSORT算法计算每个样本中不同免疫细胞的占比，进行免疫细胞浸润分析和Kaplan-Meier生存分析。通过LASSO回归、单因素和多因素Cox回归筛选HNSCC样本中免疫细胞并构建免疫细胞风险评分预后模型，采用Kaplan-Meier法和ROC曲线评估该模型。结果：在22种TIICs中，M0、M1和M2巨噬细胞具有相对高的百分比，约占40%。CD8+T细胞与活化CD4+记忆T细胞呈显著正相关(r = 0.55)，而与M0巨噬细胞(r = −0.53)呈显著负相关。生存分析表明高表达的调节性T细胞(P < 0.05)和滤泡辅助性T细胞(P < 0.05)预后较好。多因素Cox分析显示调节性T细胞(P < 0.05)为HNSCC患者预后的独立因素。预后模型中低风险组患者的总生存期明显高于高风险组(P < 0.001)。结论：TIICs与HNSCC患者预后关系密切。调节性T细胞可考虑成为治疗HNSCC的靶标，本研究构建的预后模型有望为临床预后提供参考。

Abstract: Objective: This paper aims to investigate the relationship between the degree of tumor infiltrating immune cells (TIICs) and prognosis of head and neck squamous cell carcinoma (HNSCC) patients, and construct a prognostic model based on multiple immune cells. Methods: The gene expression profiles and survival information of 515 HNSCC samples and 44 normal control samples were downloaded from the TCGA database. The proportion of different immune cells in each sample was calculated using the CIBERSORT algorithm, and immune cell infiltration analysis and Kaplan-Meier survival analysis were performed. The immune cells in HNSCC samples were screened by LASSO regression, univariate and multivariate Cox regression, and an immune cell risk scoring prognostic model was constructed. The model was evaluated by Kaplan-Meier method and ROC curve. Results: Among the 22 TIICs, M0, M1, and M2 macrophages had relatively high percentages, accounting for about 40%. CD8+T cells were significantly positively correlated with activated CD4+ memory T cells (r = 0.55), while significantly negatively correlated with M0 macrophages (r = −0.53). Survival analysis showed that high expression of regulatory T cells (P < 0.05) and follicular helper T cells (P < 0.05) had better prognosis. Multivariate Cox analysis showed that regulatory T cells (P < 0.05) were an independent prognostic factor for HNSCC patients. The total survival period of the low-risk group was significantly longer than that of the high-risk group (P < 0.001). Conclusion: TIICs are closely related to the prognosis of HNSCC patients. Regulatory T cells may be considered as a target for treatment of HNSCC, and the prognostic model constructed in this study may provide reference for clinical prognosis.