Assessment of causal association between autoimmune thyroiditis and thyroid cancer: A Mendelian randomization study.

Abstract

Currently, the precise interplay between autoimmune thyroiditis, particularly Hashimoto thyroiditis, and thyroid cancer remains ambiguous. While certain observational studies suggest autoimmune thyroiditis (including Hashimoto thyroiditis) as a predisposing factor for thyroid cancer. Nevertheless, it is still uncertain whether autoimmune thyroiditis is independently associated with thyroid cancer. We employed Mendelian randomization (MR) study methodology, a genetic analysis approach, to evaluate the causal impact of autoimmune thyroiditis on the occurrence of thyroid cancer. We obtained and synthesized statistical data by utilizing public available genome-wide association studies (GWAS). Our study utilized GWAS summary statistics datasets associated with autoimmune thyroiditis (including Hashimoto thyroiditis) as the exposure data source and selected GWAS summary statistics datasets related to thyroid cancer as the outcome data source. Single nucleotide polymorphisms closely associated with autoimmune thyroiditis were chosen as instrumental variables. We conducted 2-sample MR analyses to elucidate the causal association between autoimmune thyroiditis and thyroid cancer. The inverse variance-weighted (IVW) method was employed as the primary methodology, supplemented by additional MR methods including MR-Egger regression, weighted median, simple mode, and weighted mode analyses, to bolster the robustness of our findings. The MR analysis conducted using the IVW method did not confirm a causal relationship between autoimmune thyroiditis and thyroid cancer (odds ratio [OR] = 0.8554, 95% confidence interval [CI]: 0.7193 to 1.0172, P = .0772; OR = 0.8477, 95% CI: 0.7159 to 1.0039, P = .0555; and OR = 1.1324, 95% CI: 0.9342 to 1.3725, P = .2052, from 3 eligible dataset analyses, respectively). Additionally, MR analysis did not observe a causal association between Hashimoto thyroiditis and thyroid cancer (OR = 1.0449, 95% CI: 0.9400 to 1.1615, P = .4155; and OR = 0.9897, 95% CI: 0.8174 to 1.1984, P = .9159, from 2 eligible dataset analyses, respectively). Consistency in results across alternative MR methods was observed. This study employing MR methodology indicates the absence of significant causal relationship between exposure to autoimmune thyroiditis (including Hashimoto thyroiditis) and thyroid cancer. Further validation through larger-scale studies with increased sample sizes is warranted in future investigations.