Study On The Causal Associations Between Thyroid Function And Lipid Metabolism-related Traits

Dyslipidemia mainly includes elevated serum total cholesterol(TC),low-density lipoprotein cholesterol(LDL-C)and triglyceride(TG)or decreased high-density lipoprotein cholesterol(HDL-C),which is a common chronic disease and an important and independent risk factor for various diseases,besides,the prevalence of dyslipidemia varies in different regions of the world but is generally high.Therefore,understanding the causes of elevated blood lipid levels is of great public health significance for the effective prevention and treatment of abnormal lipid metabolism,reducing the burden of related diseases,and improving quality of life.Observational studies have shown that hypothyroidism is an important risk factor for dyslipidemia,and thyroid function status is closely associated with lipid metabolism.However,the methods and sample sizes of these epidemiological studies are different,and there are certain differences in the research results.Furthermore,observational studies are susceptible to measurement error,confounding factors,as well as reverse causality and it is difficult to determine whether a causal relationship exists between thyroid function and lipid metabolismrelated traits.Mendelian randomization(MR)serves as an efficient and cost-effective method to make causal inference in observational studies.Our study combined MR,genetic correlation,and colocalization analysis to explore the causal relationship between thyroid function and lipid metabolism-related traits by using largescale genome-wide associated study(GWAS)summary data from European ancestry.Thyroid function indicators included normal range free thyroxine(FT4),normal range thyroidstimulating hormone(TSH),hypothyroidism and hyperthyroidism,and the GWAS summary data were obtained from the ThyroidOmics Consortium.Lipid metabolism-related traits included high-density lipoprotein cholesterol(HDL-C),low-density lipoprotein cholesterol(LDL-C),total cholesterol(TC),triglyceride(TG),apolipoprotein A1(ApoA1),apolipoprotein B(ApoB),lipoprotein A(Lp(a))and hypercholesterolaemia,of which HDL-C,LDL-C,TC,and TG data were from the Global Lipids Genetics Consortium,as well as ApoA1,ApoB,Lp(a)and hypercholesterolaemia were from the UK Biobank.In terms of analysis strategy,up to seven MR methods based on different modeling assumptions were used to verify the consistency of the results;and genetic correlation analysis was used to tested whether the two complex traits had a similar genetic structure and whether pleiotropy would influence the result of MR;as well as colocalization analysis was employed to examined whether two traits had shared causal genetic variation within a given gene region to rule out genetic confounding due to linkage disequilibrium between SNPs.Our study not only advance the application of genomics data to develop a more comprehensive etiological exploration of complex traits in actual data analysis,but also provides effective intervention strategies and guidance for the prevention of abnormal lipid metabolism in clinical work.Result:1.Genetic correlation analysis showed that there was no significant genetic correlation between thyroid function and lipid metabolism-related traits.Overall,the possibility of pleiotropy between thyroid function and lipid metabolism-related traits was low.2.The results of MR suggested that normal FT4 had a negative causal association with LDL-C,TC and ApoB;as well as normal TSH had a causal relationship with LDL-C,TC and hypercholesterolaemia.There were no significant causal associations between hypothyroidism and lipid metabolism-related traits,hyperthyroidism and lipid metabolism-related traits.Bidirectional MR showed that lipid metabolism-related traits had no significant causal effects on thyroid function,ruling out the possibility of reverse causal association.Several MR methods were used to minimize the impact of model misspecifications on the results and different methods yielded consistent estimates of causal effects,indicating that these results were credible.3.Co localization analysis suggested that there existed a causal genetic variant between normal FT4 and LDL-C,two causal genetic variants between normal FT4 and TC,two causal genetic variants between normal FT4 and ApoB,two causal genetic variants between normal TSH and LDL-C,two causal genetic variants between normal TSH and TC,confirming the existence of causal relationships.Conclusion:1.The present study investigated the causal relationships between thyroid function and lipid metabolism-related traits based on large-scale GWAS summary statistics.We found that normal FT4 was causally associated with LDL-C,TC,and ApoB;as well as normal TSH was causally associated with LDL-C,TC and hypercholesterolemia.These results suggested that even in euthyroid patients,the changes of FT4 or TSH levels could significantly affect the levels of lipid metabolism-related traits,which provided the remainder that people should keep FT4 at a normal-high level or TSH at a normal-low level to reduce the risk of abnormal lipid metabolism.2.Genetic correlation,MR and colocalization analysis illustrated the relationship between complex traits from different perspectives.In data analysis studies,various methods can be used to obtain more convincing findings.