Abstract
PURPOSE: Our insight in the genetics of Hashimoto's thyroiditis (HT) has become clearer through information provided by genome-wide association studies and candidate gene studies, but remains still not fully understood. Our aim was to assess how many different genetic risk variants contribute to the development of HT. METHODS: 147 HT cases (10.2% men) and 147 controls (13.6% men) were qualified for the analysis. Intrinsic and environmental factors were controlled for. Polymorphisms (SNP) were chosen based on the literature and included markers of the genes PTPN22, CTLA4, TG, TPO among others, and of genomic regions pointed by GWAS studies. SNP were typed on a microarray. Variants in the HLA-DRB1 gene were identified by Sanger sequencing. RESULTS: Multivariate predisposition to HT was modeled. Based on the investigated group, a model of seven variables was obtained. The variability explained by this model was assessed at only 5.4821% (p = 2 × 10(-6)), which indicates that many dozens of factors are required simultaneously to explain HT predisposition. CONCLUSIONS: We analyzed genetic regions commonly and most significantly associated with autoimmune thyroid disorders in the literature, on a carefully selected cohort. Our results indicated a lack of possibility to predict the risk of HT development, even with a multivariate model. We therefore conclude that strong associations of single genetic regions with HT should be interpreted with great caution. We believe that a change in the attitude towards genetic association analyses of HT predisposition is necessary. Studies including multiple factors simultaneously are needed to unravel the intricacies of genetic associations with HT.