Scholarlink

Background <p>Telomeres, located at chromosome ends, regulate cell division and maintain genomic stability. Telomere-related genes (TRGs) play essential roles in tumorigenesis and immune escape, but their mechanisms in lung adenocarcinoma (LUAD) are not fully understood.</p> Methods <p>This study integrated genomics, transcriptomics and pathological images to analyze TRGs expression in LUAD. Univariate Cox regression identified 278 TRGs linked to prognosis. Using consensus clustering, 1086 LUAD patients were classified into two distinct molecular subtypes (Cluster 1 and Cluster 2). Reliability was validated through non-negative matrix factorization (NMF) and nearest template prediction (NTP), and deep learning models (CLAM, TransMIL, CAMIL) were constructed to predict these molecular subtypes from whole-slide images (WSIs).</p> Results <p>Our study found that TRG activation is associated with the occurrence, progression, and poor prognosis of LUAD. TRG expression-based stratification identified subtypes with distinct clinical outcomes. Specifically, Cluster 1 showed a higher frequency of KRAS mutations, together with high microsatellite instability (MSI), increased immune infiltration, and activation of the immune cycle, along with a better response to immune checkpoint blockade (ICB) therapy and more favorable patient prognosis. In contrast, Cluster 2 was characterized by high tumor mutational burden (TMB) and homologous recombination deficiency (HRD). Although these genomic features would be expected to enhance tumor immunogenicity, Cluster 2 exhibited profound adaptive immune resistance, including marked T-cell exhaustion (elevated LAG3 and TIGIT) and a profoundly immunosuppressive tumor microenvironment (TME). Microbial profiling further revealed that the genera Blautia, Faecalibacterium, and Leuconostoc were significantly enriched in Cluster 1. Finally, we developed a deep learning model that accurately predicts TRG-based subtypes, providing preliminary support for the potential clinical utility of this classification framework.</p> Conclusion <p>This study clarifies the association between TRGs expression and the biological characteristics of LUAD, highlighting their potential clinical significance in molecular typing and therapeutic guidance, and laying a foundation for personalized diagnosis and treatment strategies for LUAD.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Characterization of telomere-related gene subtypes in lung adenocarcinoma and their implications for prognosis and treatment

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月11日

查看全文
  • Jingyi Gao,
  • Yuansheng Zhao,
  • Zhiying Cheng,
  • Jiajun Yang,
  • Linjun Jiang,
  • Shuxue Xi,
  • Shufang Shi,
  • Geng Tian,
  • Haiwen Zhao,
  • Jialiang Yang,
  • Jinyang Liu

摘要

Background

Telomeres, located at chromosome ends, regulate cell division and maintain genomic stability. Telomere-related genes (TRGs) play essential roles in tumorigenesis and immune escape, but their mechanisms in lung adenocarcinoma (LUAD) are not fully understood.

Methods

This study integrated genomics, transcriptomics and pathological images to analyze TRGs expression in LUAD. Univariate Cox regression identified 278 TRGs linked to prognosis. Using consensus clustering, 1086 LUAD patients were classified into two distinct molecular subtypes (Cluster 1 and Cluster 2). Reliability was validated through non-negative matrix factorization (NMF) and nearest template prediction (NTP), and deep learning models (CLAM, TransMIL, CAMIL) were constructed to predict these molecular subtypes from whole-slide images (WSIs).

Results

Our study found that TRG activation is associated with the occurrence, progression, and poor prognosis of LUAD. TRG expression-based stratification identified subtypes with distinct clinical outcomes. Specifically, Cluster 1 showed a higher frequency of KRAS mutations, together with high microsatellite instability (MSI), increased immune infiltration, and activation of the immune cycle, along with a better response to immune checkpoint blockade (ICB) therapy and more favorable patient prognosis. In contrast, Cluster 2 was characterized by high tumor mutational burden (TMB) and homologous recombination deficiency (HRD). Although these genomic features would be expected to enhance tumor immunogenicity, Cluster 2 exhibited profound adaptive immune resistance, including marked T-cell exhaustion (elevated LAG3 and TIGIT) and a profoundly immunosuppressive tumor microenvironment (TME). Microbial profiling further revealed that the genera Blautia, Faecalibacterium, and Leuconostoc were significantly enriched in Cluster 1. Finally, we developed a deep learning model that accurately predicts TRG-based subtypes, providing preliminary support for the potential clinical utility of this classification framework.

Conclusion

This study clarifies the association between TRGs expression and the biological characteristics of LUAD, highlighting their potential clinical significance in molecular typing and therapeutic guidance, and laying a foundation for personalized diagnosis and treatment strategies for LUAD.