Transcriptome-wide gene-age interaction study and multi-omics comprehensive analysis reveals heterogeneous regulation...

Substantial age-related heterogeneity exists in the treatment of head and neck squamous cell carcinoma (HNSCC) patients, effective predictive guidance is needed to improve survival of elderly HNSCC patients, but it is unclear whether the effect of gene expression on HNSCC overall survival varies with age. By integrating transcriptome data from multiple centers, we identified one gene, NLRC4, whose effect was significantly modified by age on a transcriptome-wide scale (HR_TCGA = 1.025, P = 8.50 × 10^-05, FDR-q = 8.78 × 10^-03; HR_GEO = 1.035, P = 2.39 × 10^-06, FDR-q = 1.41 × 10^-02; HR_combined = 1.029, P = 4.96 × 10^-09, HR_ArrayExpress = 1.052, P = 1.51 × 10^-02). Moreover, there was an antagonistic interaction between NLRC4 low expression and aging. Comprehensive analysis using multi-omics data suggested that immune-related indicators showed varying degrees of differences in NLRC4-age subgroups, NLRC4 was highly expressed in monocytes/macrophages, and may associated with M1/M2 macrophage polarization. Subsequent transcriptome-wide NLRC4-age-based three-way interaction analysis reflected the broad association among NLRC4, age and ECM-related genes (mainly the collagen family) on the prognosis of HNSCC. Identified three-way interactions can better improve prognostic accuracy than two-way interactions. Our research identified the complex interaction effects among multiple environment and transcriptomic factors, and provided therapeutic target for HNSCC patients.