表观遗传学分析揭示甲状腺癌中的异常衰老

Thyroid cancer (TC), the most common endocrine malignancy, is closely linked to aging. In this study, we estimated DNA methylation (DNAm) age in TC tissues versus adjacent non-cancerous thyroid tissues using TCGA and GSE97466 datasets.

We employed multi-tissue DNAm age estimators, specifically Horvath's clock and the Bayesian Neural Network (BNN) clock, to quantify epigenetic aging in TC and adjacent normal tissues from the TCGA and GSE97466 cohorts. We further integrated genomic, transcriptomic, and mutational analyses, including assessments of senescence-associated secretory phenotype (SASP), immune cell infiltration, and thyroid differentiation, to elucidate the associations between DNAm age acceleration and various cancer-related features.

Our results revealed marked DNAm age acceleration in TC tissues relative to controls, with both models yielding significantly higher DNAm age estimates in tumors. Although DNAm age correlated strongly with chronological age in control tissues ( R ≈ 0.94), this association was notably diminished in TC tissues ( R ≈ 0.62). Moreover, TC tissues with accelerated DNAm age exhibited higher telomerase-associated gene expression score, heightened SASP signaling, and distinct immune profiles characterized by elevated M2 macrophages and reduced activated NK cells. Genomic analysis further showed a higher prevalence of BRAF mutations and lower thyroid differentiation scores (TDS) in the accelerated subgroup. No significant associations were detected between DNAm age and clinicopathological parameters or overall survival.

Our findings provide valuable insights into the complex interplay between epigenetic regulation, immune dysfunction, and genomic instability in thyroid cancer, highlighting promising avenues for future research and therapeutic intervention.