ABSTRACT

Coffee consumption has been previously reported to be linked to the incidence of thyroid cancer. However, the exact nature of the causal relationship between the 2 remains elusive. We conducted a 2-sample Mendelian randomization (MR) study to evaluate the relationship between coffee consumption and thyroid cancer incidence. We used 36 SNPs associated with coffee consumption obtained from genome-wide association study (GWAS) summary data (ieu-a-1082) involving 212,119 female UK Biobank subjects with White British ancestry. Thyroid cancer incidence was estimated using the public GWAS summary data and the FinnGen research project. Sensitivity analysis was conducted with random-effects inverse-variance weighted MR analysis to assess the effects of possible MR assumption violations. The GWAS summary data (ieu-a-1082) did not reveal a causal relationship between SNPs related to coffee consumption and thyroid cancer incidence. The summary statistics based on the FinnGen research project and obtained using the inverse-variance weighted method revealed no markedly increased coffee consumption risk among patients with different thyroid cancers: follicular thyroid cancer (odds ratio [OR] = 1.101, 95% confidence interval [CI]: 0.492-2.467, P = .814), thyroid malignancy (OR = 0.915, 95% CI: 0.722-1.159, P = .460), and papillary thyroid adenocarcinoma (OR = 0.938, 95% CI: 0.716-1.227, P = .640). Similarly, MR-Egger, weighted median, simple mode, and weighted mode methods produced consistent findings. No significant bias, heterogeneity, or pleiotropy was detected. The large-scale MR analysis demonstrated that although genetically predicted coffee intake is not linked to thyroid cancer incidence, a weak relationship may not be excluded.

‍