Abstract
BACKGROUND: Thyroid cancer, the most common malignancy in the endocrine system, has seen a global increase in incidence. This study aims to investigate the causal relationship between serum circulating microRNAs (miRNAs) and thyroid cancer risk using a Mendelian randomization (MR) approach. METHODS: We conducted a two-sample MR analysis using miRNA expression quantitative trait loci (eQTL) data from two independent cohorts and thyroid cancer genome-wide association study (GWAS) data. The discovery cohort included miRNA expression levels quantified via qRT-PCR in whole blood samples, while the validation cohort comprised miRNA expression data from blood samples of unrelated individuals of European ancestry. We applied various MR methods, including inverse variance-weighted (IVW) and MR-Egger, to assess the genetic associations between miRNAs and thyroid cancer. Additionally, we performed target gene prediction and pathway analysis to explore the biological mechanisms underlying the observed associations. RESULTS: Our analysis identified three miRNAs significantly associated with thyroid cancer risk: miR-hsa-125b-5p and miR-hsa-30a-3p were found to have harmful effects, while miR-hsa-130a-3p exhibited a protective effect. These findings were consistently validated across both cohorts. The target genes of these miRNAs were enriched in pathways related to gland development, myeloid cell differentiation, cellular senescence, FoxO signaling pathway, and p53 signaling pathway, providing insights into the potential molecular mechanisms linking these miRNAs to thyroid cancer. CONCLUSION: This study suggests potential causal associations between specific serum circulating miRNAs and thyroid cancer risk using a Mendelian randomization approach. The identified miRNAs, including miR-hsa-125b-5p, miR-hsa-30a-3p, and miR-hsa-130a-3p, could be further investigated as potential biomarkers for risk prediction and early diagnosis of thyroid cancer.