Abstract
This study explores the potential causal relationship between bioavailable testosterone (BioT) and thyroid carcinoma using 2-sample Mendelian randomization. The research data were collected from the genome-wide association studies (GWASs). Two sets of genetic variants data of BioT from the UK Biobank (male cohort contains 1,78,782 samples and female cohort contains 1,88,507 samples). Thyroid carcinoma outcomes were selected from the FinnGen GWAS (989 thyroid carcinoma cases and 2,17,803 control cases). One hundred forty-nine and 96 single nucleotide polymorphisms (SNPs) closely related to BioT were selected as IVs in female cohort and male cohort respectively. Mendelian randomization (MR) was conducted by inverse variance weighted (IVW), weighted median and MR-Egger. At the same time, MR-Egger and Mendelian randomization-pleiotropy residual sum and outlier were used to detect horizontal pleiotropy, and leave-one-out sensitive analysis was conducted. Using the IVW approach, we found BioT was suggestively negatively correlated with the risk of thyroid carcinoma in female cohort (odds ratios = 0.690, 95% confidence interval = 0.482-0.988, P = .043), but there was no significant correlation in the male cohort (P = .461). Potential heterogeneity and horizontal pleiotropy were not observed in sensitivity analysis. Our findings suggested that BioT may have a causal relationship with thyroid carcinoma, which could potentially be a protective factor for female thyroid carcinoma. Further studies are needed to confirm these preliminary findings.