BMC Medical Genomics (May 2024)

Identification and validation of a novel risk model based on cuproptosis‑associated m6A for head and neck squamous cell carcinoma

  • Zhongxu Xing,
  • Yijun Xu,
  • Xiaoyan Xu,
  • Kaiwen Yang,
  • Songbing Qin,
  • Yang Jiao,
  • Lili Wang

DOI
https://doi.org/10.1186/s12920-024-01916-5
Journal volume & issue
Vol. 17, no. 1
pp. 1 – 21

Abstract

Read online

Abstract Background Head and neck squamous cell carcinoma (HNSCC) is a prevalent cancer with a poor survival rate due to anatomical limitations of the head and a lack of reliable biomarkers. Cuproptosis represents a novel cellular regulated death pathway, and N6-methyladenosine (m6A) is the most common internal RNA modification in mRNA. They are intricately connected to tumor formation, progression, and prognosis. This study aimed to construct a risk model for HNSCC using a set of mRNAs associated with m6A regulators and cuproptosis genes (mcrmRNA). Methods RNA-seq and clinical data of HNSCC patients from The Cancer Genome Atlas (TCGA) database were analyzed to develop a risk model through the least absolute shrinkage and selection operator (LASSO) analysis. Survival analysis and receiver operating characteristic (ROC) analysis were performed for the high- and low-risk groups. Additionally, the model was validated using the GSE41613 dataset from the Gene Expression Omnibus (GEO) database. GSEA and CIBERSORT were applied to investigate the immune microenvironment of HNSCC. Results A risk model consisting of 32 mcrmRNA was developed using the LASSO analysis. The risk score of patients was confirmed to be an independent prognostic indicator by multivariate Cox analysis. The high-risk group exhibited a higher tumor mutation burden. Additionally, CIBERSORT analysis indicated varying levels of immune cell infiltration between the two groups. Significant disparities in drug sensitivity to common medications were also observed. Enrichment analysis further unveiled significant differences in metabolic pathways and RNA processing between the two groups. Conclusions Our risk model can predict outcomes for HNSCC patients and offers valuable insights for personalized therapeutic approaches.

Keywords