Oxidative Stress Links Thyroid Autoimmunity to Cancer: Peroxiredoxin 2 Protection via Genomic and Single-Cell Insights

Objective: This study identifies shared genetic factors linking Hashimoto's thyroiditis (HT) and thyroid cancer (TC) using an integrated multiomics approach. Methods: We combined Mendelian randomization (MR) analysis using FinnGen genome-wide association study data, single-cell RNA sequencing of 76,243 thyroid cells, and machine learning classification models to identify causal genes and their expression patterns across disease states. Results: MR analysis identified 10 genes with consistent directional effects across both diseases. Peroxiredoxin 2 (PRDX2) emerged as the strongest protective factor (HT: odds ratio [OR] = 0.54, 95% confidence interval [CI]: 0.31-0.94; TC: OR = 0.68, 95% CI: 0.50-0.91). Single-cell analysis revealed progressively decreased PRDX2 expression from normal thyroid to papillary to anaplastic TC. Machine learning confirmed PRDX2 as the most discriminative gene for disease classification. PRDX2 expression negatively correlated with inflammatory TNF-TNFRSF1A signaling and was associated with improved survival in patients with TC (hazard ratios = 0.33, 95% CI: 0.11-0.96, p = 0.043). Conclusions: PRDX2 functions as a key protective factor in both HT and TC pathogenesis, likely through modulation of oxidative stress and inflammatory signaling. These findings provide mechanistic insights into the HT-TC relationship and highlight PRDX2 as a promising therapeutic target for thyroid diseases.