The MR method was utilized to explore the bidirectional causal association between RA and thyroid dysfunction. In the European population, we identified a reciprocal causal relationship between RA and hypothyroidism. Specifically, we found that the genetic predisposition to RA is linked to an elevated risk of hypothyroidism, and conversely, individuals with hypothyroidism have an increased susceptibility to RA. Furthermore, we observed a positive causal association between RA and hyperthyroidism. Nevertheless, the reverse MR analysis yielded inconclusive results, suggesting that there is no significant causal relationship between hyperthyroidism and RA.
From our observations in clinical practice, we have identified thyroid disease as a prevalent comorbidity in patients diagnosed with RA. Previous observational studies demonstrated that individuals with RA who also had hypothyroidism, increased disease activity and joint tenderness, and correction of hypothyroidism in individuals with RA can significantly improve the disease activity of patients with RA[26]. Likewise, a prospective cohort study revealed that female patients with RA had a three-fold higher incidence of hypothyroidism compared to the general population[27]. This co-occurrence of hypothyroidism in female RA patients was found to be associated with a heightened risk of cardiovascular diseases (CVD)[27]. Recently, a meta-analysis from China determined that individuals diagnosed with RA exhibited a heightened susceptibility to developing thyroid disorders, especially hypothyroidism, which was 2.25 times higher than non-RA patients; And the increased risk of hyperthyroidism was only secondary, which was 1.65 times higher than that of non-RA patients[28]. However, these studies could not explain a causal effect between RA and thyroid dysfunction, they provided sufficient evidence for an association between RA and hyperthyroidism and hypothyroidism. Through Mr.'s research, we have demonstrated that RA may increase the incidence of hyperthyroidism and hypothyroidism, and hypothyroidism may also increase the incidence of RA, which strengthens and enriches the findings of these previous observational investigations.
Mendelian randomization (MR) employs genetic variations that have strong associations with a specific exposure as instrumental variables to assess the causal effects of the exposure on diverse study outcomes[16, 29]. In our MR study, we initially chose SNPs that exhibited strong associations with the exposure factors. The selection of these SNPs was based on the three fundamental hypotheses of association, independence, and exclusivity. Five different analysis methods are used for MR analysis. The utilization of the IVW method to detect RA revealed an elevated relative risk of hyperthyroidism (OR = 1.33, 95% CI = 1.17–1.52, P = 2.407e-05), as well as a heightened risk of hypothyroidism (OR = 1.29, 95% CI: 1.21–1.37, P = 3.614e-16). On the flip side, it was observed that hypothyroidism might also elevate the relative risk of developing rheumatoid arthritis (OR = 1.57, 95% CI = 1.30–1.91, P = 4.211e-06). The results deduced by the other four methods accorded with the IVW analysis method. We also performed a sensitivity analysis, and heterogeneity testing revealed significant heterogeneity between each instrumental variable. However, it does not affect the final analysis results. It is worth noting that the MR Egger method suggests that multiple IVs do not have horizontal pleiotropy, so the research results may be less affected by gene pleiotropy.Through the above methods, we ultimately consider that the genetic susceptibility of RA is linked to a higher chance of developing hypothyroidism, and vice versa. The genetic susceptibility of RA is linked to a higher chance of developing hyperthyroidism, but hyperthyroidism seems to have no causal effect on RA. Graves' disease (GD) is a prevalent cause of hyperthyroidism, and research has indicated that 16.7% of individuals with GD also have another autoimmune disease[30]. Among these cases, rheumatoid arthritis (RA) has a prevalence rate of 1.9%[30]. In an Asian population, a Mendelian randomization study revealed a reciprocal causal relationship between GD and RA[31]. Due to the possibility that Mendelian randomization analysis may be influenced by factors such as race or region, further research is needed on the causality between hyperthyroidism and RA based on different races or regions.
Additionally, the association between RA and thyroid diseases may have some other potential pathogenic mechanisms. Hyperthyroidism and hypothyroidism are mainly caused by AITD. There exists a correlation between RA and autoimmune thyroid diseases (AITD), yet the underlying causes of their co-occurrence remain unclear. Genetic variations in the human leukocyte antigen (HLA) system, cytotoxic T lymphocyte-associated protein 4 (CTLA-4), and protein tyrosine phosphatase non-receptor 22 (PTPN22) genes have been found to be linked to an increased susceptibility to autoimmune thyroid diseases (AITD)[8, 32, 33]. CTLA4 plays a negative regulatory role in T lymphocyte immune response[34]. Common allelic variations in CTLA-4 expression levels are major determinants of susceptibility to autoimmune diseases such as autoimmune thyroid disease and RA[32]. PTPN22 is associated with TCR signaling and participates in regulating CBL function in T cell receptor signaling pathways[35]. Mutations in this gene can alter TCR regulation and T cell activation, closely related to the increased risk of various autoimmune diseases, including RA and GD. Recent findings have indicated that the expression of CXCL10 significantly rises in serum and/or tissues of organ-specific autoimmune diseases like GD and RA[10]. The production of IFN-γand TNF-αin thyroid tissue may be linked to the recruitment of Th1 lymphocytes, triggering the release of CXCL10 by thyroid cells, binding to CXCR3 on the surface of Th1 lymphocytes, and promoting the recruitment of Th1 lymphocytes into the thyroid, secreting IFN-γ and TNF-α, thus forming an amplified feedback loop, resulting in the persistence of the disease[36–37]. Similarly, many chemokines, including CXCL10, are present in the joints of RA, promote lymphocyte recruitment into the synovium, and are involved in germinal center formation[38]. Therefore, CXCL10 can be used as a therapeutic target to prevent immune cells from recruiting to tissues, which needs to be further studied, and provides a new therapeutic strategy for RA and thyroid disease comorbidity.
Our research has some advantages. Firstly, the present analysis employing MR is grounded in GWAS, leveraging genetic data for causal inference purposes. Compared with traditional epidemiological methods, MR reduces the impact of confounding factors, reverse causality, and ethical ethics. Secondly, we employed diverse approaches for Mendelian randomization analysis and sensitivity analysis to ensure the robustness and validity of the final outcomes. Significantly, we have established a reciprocal causal association between RA and hypothyroidism, offering novel perspectives for clinicians in terms of RA screening and management strategies.
Certainly, there are certain limitations in this study. Firstly, the inclusion of SNPs in this study was limited to European populations, thus requiring further confirmation to establish the causal relationship between RA and hyperthyroidism and hypothyroidism in other racial or regional contexts. Secondly, due to the limitation of GWAS pooled data, it is impossible to perform stratified analysis according to factors such as disease course and severity.
Conclusions
The evidence we provide suggests that the genetic susceptibility of RA is linked to the increasing risk of hyperthyroidism and hypothyroidism. Conversely, it is possible that genetic predisposition to hypothyroidism is linked to a higher risk of developing RA, while there may not be a causal association between genetic susceptibility to hyperthyroidism and an increased risk of RA. As a result, it is crucial to enhance screening and management of thyroid disorders in individuals with RA to enhance patient prognosis.