Integrative Analysis of Telomere-Related Genes Reveals Prognostic Signatures in Laryngeal Cancer

Abstract

Background

Laryngeal cancer (LC) is a common malignant tumor. Telomere-related genes (T-RGs) play critical roles in cellular senescence and carcinogenesis, but their prognostic relevance in LC remains to be fully elucidated. Therefore, exploring the prognostic genes related to telomeres in LC is important.

Methods

Public retrospective datasets TCGA-HNSC and T-RGs were used to identify candidate genes by intersecting differentially expressed genes with T-RGs. Key analytical approaches, including machine learning algorithms and univariate Cox regression, were applied to identify prognostic genes and construct a prognostic model. A nomogram was developed to assess the prognostic value for LC based on overall survival. Disease samples were classified into high-risk and low-risk groups, and subsequent analyses included immune infiltration, immune checkpoint expression, and related evaluations. Experimental validation of prognostic genes was performed through RT-qPCR.

Results

A total of 314 candidate genes were obtained from 8961 differentially expressed genes. Four key prognostic genes (CHTF18, FANCG, NR5A1, and XRCC3) were identified. The constructed risk score model retained consistent predictive accuracy in both the training and validation datasets, with AUCs ranging from approximately 0.61 to 0.9. Enriched activated immune cells were detected in the low-risk group through immune microenvironment analysis, whereas immune suppression–related features were identified in the high-risk group, accompanied by a reduced tumor mutational burden that was detected. Finally, preliminary validation using RT-qPCR in a limited cohort of clinical samples indicated that the expression trends of three prognostic genes were elevated in LC tissues, showing concordance with the bioinformatic findings.

Conclusions

This study identified four key prognostic T-RGs (CHTF18, FANCG, NR5A1, and XRCC3) and constructed a corresponding prognostic model for LC. Our analyses further suggest a potential link between telomere maintenance mechanisms and the tumor immune microenvironment, which may influence patient outcomes.