A Mendelian Randomization Study Of The Causal Association Between Testosterone And Lipid Profiles In Men

Background:Testosterone is a steroid hormone that plays an essential role in male growth,maintenance of libido,reproductive function,and aging.Male hypogonadism and lack of testosterone can promote the aging of internal organs,increase the prevalence of systemic diseases such as metabolic syndrome,cardiovascular disease,and Alzheimer’s disease,and seriously affect the quality of life.Dyslipidemia is characterized by an adverse lipid profile that includes elevated levels of total cholesterol(TC),low-density lipoprotein cholesterol(LDL),triglycerides(TG)and often associated with lower levels of high-density lipoprotein cholesterol(HDL)The combination.Due to the changes in Chinese lifestyle in recent years,the prevalence of hyperlipidemia is increasing year by year,and the relationship between testosterone and dyslipidemia in men is well documented in observational studies.The relationship between testosterone and blood lipids is complex and affected by many factors.The effect of endogenous testosterone on blood lipids varies with age,environment,nutritional status,and gender.The effect of exogenous testosterone on blood lipids also varies with the preparations,application methods and types of diseases treated.Testosterone can regulate the function of the cardiovascular system by affecting lipid metabolism,and affect the occurrence and development of atherosclerosis.Therefore,it is of great significance to explore the effect of male testosterone on blood lipid profile.However,all of these findings are drawn from observational studies and are prone to reverse causation,selection bias,and especially unobserved confounders,correlation cannot simply be equated with causation.Before translating these findings into clinical practice,further investigation of the causality behind these associations is necessary.Two-sample Mendelian randomization(TSMR)provides an alternative method to infer causality in observational studies by using genetic tools(single nucleotide polymorphisms,SNPs).It utilizes random assignment of genetic alleles affecting exposure and largely excludes the confounding of unobserved confounders and avoids reverse causality compared to other study designs.Objectives:Observational studies have shown that male testosterone is negatively correlated with TC,LDL,and TG,and positively correlated with HDL.But it was unclear whether the association was causal.This study used the TSMR method to explore the causal relationship and magnitude of male testosterone and dyslipidemia.Methods:This study used the currently publicly searchable Genome-Wide Association Studies(GWAS)database to search for relevant data on testosterone and lipid profiles in men.Select the instrumental variables(Instrumental Variables,IVs)that meet the target exposure group,and retrieve data according to the selected instrumental variables in the corresponding outcome,namely the lipid profile database.Pooled analyses of exposure group IVs(testosterone in men)and outcome data(lipid profiles)were performed.Using 141(HDL),142(LDL),139(TC),and 138(TG)SNPs associated with male testosterone as instrumental variables,data from the GWAS Catalog were used to evaluate the causal association between male testosterone and dyslipidemia.Finally,the results are presented in the form of scatter plots(inverse variance weighted model,IVW);pleiotropic tests are presented in the form of statistical tables(MRegger model);quality control is presented in the form of forest plots(weighted median method,MR-raps,MR-Lasso,MR-Robust,MR-Egger,MR-PRESSO);Heterogeneity tests were validated using Q-test(IVW model);the final sensitivity analysis is presented in the form of a forest plot(leave one out).Results:Finally,HDL,LDL,TC and TG had 141,142,139 and 138 SNPs as instrumental variables,respectively.IVW model results showed that male testosterone was causally associated with dyslipidemia,with a standard deviation increase of 0.056 in high-density lipoprotein(HDL;P=0.006)for each standard deviation increase in testosterone(TTM)in normal men.There was no causal relationship between TTM and triglycerides(TG),total cholesterol(TC),and lowdensity lipoprotein cholesterol(LDL).We have used several different MR methods for validation with consistent results.Our study shows that there is indeed a causal relationship between higher endogenous testosterone levels and elevated HDL.Conclusions:In conclusion,male testosterone is an independent risk factor for dyslipidemia,and there is a causal association between the two.The standard deviation of high-density lipoprotein increases with the increase of male total testosterone,but there is no significant causal association with triglyceride,total cholesterol,and low density lipoprotein cholesterol.This study provides evidence for a causal relationship between testosterone and risk of dyslipidemia in men.However,the results still need to be validated with large-scale intervention studies.