Table3_A Novel Immune-Gene Pair Signature Revealing the Tumor Microenvironment Features and Immunotherapy Prognosis of Muscle-Invasive Bladder Cancer.XLSX

Immunotherapy has been a milestone for muscle-invasive bladder cancer (MIBC), but only a small portion of patients can benefit from it. Therefore, it is crucial to develop a robust individualized immune-related signature of MIBC to identify patients potentially benefiting from immunotherapy. The current study identified patients from the Cancer Genome Atlas (TCGA) and immune genes from the ImmPort database, and used improved data analytical methods to build up a 45 immune-related gene pair signature, which could classify patients into high-risk and low-risk groups. The signature was then independently validated by a Gene Expression Omnibus (GEO) dataset and IMvigor210 data. The subsequent analysis confirmed the worse survival outcomes of the high-risk group in both training (p < 0.001) and validation cohorts (p = 0.018). A signature-based risk score was proven to be an independent risk factor of overall survival (p < 0.001) and could predict superior clinical net benefit compared to other clinical factors. The CIBERSORT algorithm revealed the low-risk group had increased CD8+ T cells plus memory-activated CD4+ T-cell infiltration. The low-risk group also had higher expression of PDCD1 (PD-1), CD40, and CD27, and lower expression of CD276 (B7-H3) and PDCD1LG2 (PD-L2). Importantly, IMvigor210 data indicated that the low-risk group had higher percentage of “inflamed” phenotype plus less “desert” phenotype, and the survival outcomes were significantly better for low-risk patients after immunotherapy (p = 0.014). In conclusion, we proposed a novel and promising prognostic immune-related gene pair (IRGP) signature of MIBC, which could provide us a panoramic view of the tumor immune microenvironment of MIBC and independently identify MIBC patients who might benefit from immunotherapy.

免疫治疗已成为肌层浸润性膀胱癌（muscle-invasive bladder cancer, MIBC）治疗的里程碑，但仅少数患者能从中获益。因此，构建稳健的MIBC个体化免疫相关特征以筛选潜在可从免疫治疗中获益的患者至关重要。本研究从癌症基因组图谱（The Cancer Genome Atlas, TCGA）中获取患者队列数据，从ImmPort数据库中筛选免疫相关基因，并采用优化的数据分析方法构建了由45个免疫相关基因对组成的预后特征模型，可将患者划分为高风险组与低风险组。该特征随后通过基因表达综合数据库（Gene Expression Omnibus, GEO）数据集与IMvigor210数据集完成了独立验证。后续分析证实，在训练队列（p < 0.001）与验证队列（p = 0.018）中，高风险组患者的生存结局均更差。基于该特征的风险评分被证实为总生存期（overall survival）的独立危险因素，且相较于其他临床因素，该评分可更精准地预测临床净获益。CIBERSORT算法分析显示，低风险组的CD8+ T细胞及活化记忆性CD4+ T细胞浸润水平更高。低风险组还高表达PDCD1（PD-1）、CD40与CD27，低表达CD276（B7-H3）与PDCD1LG2（PD-L2）。值得注意的是，IMvigor210数据集数据表明，低风险组的“炎型”表型占比更高，而“荒漠型”表型占比更低，且免疫治疗后低风险组患者的生存结局显著更优（p = 0.014）。综上，本研究提出了一种新颖且具有应用前景的MIBC预后免疫相关基因对（immune-related gene pair, IRGP）特征模型，该模型可全面解析MIBC的肿瘤免疫微环境，并独立筛选出可能从免疫治疗中获益的MIBC患者。