列线图
免疫系统
结直肠癌
肿瘤科
比例危险模型
免疫疗法
医学
内科学
癌变
肿瘤微环境
生存分析
免疫检查点
癌症
免疫学
作者
Leyi Zhao,Liting Xi,Yani Liu,Guoliang Wang,Ming Zong,Peng Xue,Shijie Zhu
出处
期刊:Biomedicines
[Multidisciplinary Digital Publishing Institute]
日期:2025-02-21
卷期号:13 (3): 539-539
标识
DOI:10.3390/biomedicines13030539
摘要
Background: Colorectal cancer (CRC) ranks as the third most common cancer worldwide. Tertiary lymphoid structures (TLSs), organized immune cell aggregates in non-lymphoid tissues, are linked to chronic inflammation and tumorigenesis. However, the precise relationship between TLSs and CRC prognosis remains unclear. This study aimed to develop a TLS-associated genetic signature to predict CRC prognosis and support clinical applications. Methods: Utilizing the TCGA database, we analyzed TLS-related gene expression in CRC versus normal tissues. Prognostic models were constructed using Cox and Kaplan-Meier analyses. CRC samples were stratified into high and low TLS groups via ssGSEA, with validation in the GSE75500 dataset. We identified clinical characteristics associated with TLS scores, created prognostic nomograms, analyzed the top 50 differential genes, assessed tumor mutations, estimated immune infiltration using CIBERSORT, and examined correlations between TLS scores and immune checkpoints. Results: A 13-gene TLS-associated prognostic model for CRC was developed, emphasizing immune response genes. Survival analysis indicated significantly better outcomes for the TLS-high group. Cox regression identified stage IV and M1 as independent factors influencing TLS scores. Nomogram analysis demonstrated that combining TLS scores with clinical features enhances prognostic accuracy. TLS scores were closely associated with immune checkpoint genes, suggesting potential immunotherapy benefits for TLS-high patients. Conclusions: This study developed and validated a TLS-based prognostic model for CRC, exploring relevant immune cells. The model holds promise for predicting clinical prognosis and treatment responsiveness in CRC patients.
科研通智能强力驱动
Strongly Powered by AbleSci AI