Association between genetically proxied HMGCR inhibition and male reproductive health: A Mendelian randomization study

Abstract

Background: The causal associations between statin use and male sex hormone levels and related disorders have not been fully understood. In this study, we employed Mendelian randomization for the first time to investigate these associations.

Methods: In two-sample Mendelian randomization framework, genetic proxies for hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) inhibition were identified as variants in the HMGCR gene that were associated with both levels of gene expression and low density lipoprotein cholesterol (LDL-C). We assessed the causal relationship between HMGCR inhibitor and 5 sex hormone levels/2 male-related diseases. Additionally, we investigated the association between 4 circulating lipid traits and outcomes. The "inverse variance weighting" method was used as the primary approach, and we assessed for potential heterogeneity and pleiotropy. In a secondary analysis, we revalidated the impact of HMGCR on 7 major outcomes using the summary-data-based Mendelian randomization method.

Results: Our study found a significant causal association between genetic proxies for HMGCR inhibitor and decreased levels of total testosterone (TT) (LDL-C per standard deviation = 38.7mg/dL, effect = -0.20, 95% confidence interval [CI] = -0.25 to -0.15) and bioavailable testosterone (BT) (effect = -0.15, 95% CI = -0.21 to -0.10). Obesity-related factors were found to mediate this association. Furthermore, the inhibitor were found to mediate a reduced risk of prostate cancer (odds ratio = 0.81, 95%CI = 0.7-0.93) by lowering bioavailable testosterone levels, without increasing the risk of erectile dysfunction (P = .17). On the other hand, there was a causal association between increased levels of LDL-C, total cholesterol, triglycerides (TG) and decreased levels of TT, sex hormone-binding globulin, and estradiol.

Conclusions: The use of HMGCR inhibitor will reduce testosterone levels and the risk of prostate cancer without the side effect of erectile dysfunction. LDL-C, total cholesterol, and TG levels were protective factors for TT, sex hormone-binding globulin, and estradiol.

Figure 1.

Overview of the study design. BT = bioavailable testosterone; DHEAS = dehydroepiandrosterone; E2 = estradiol; ED = erectile dysfunction; GLGC = Global Lipid Genetics Consortium; HDL = high density lipoprotein cholesterol; HEIDI = heterogeneity in dependence tool; HMGCR = 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase; LDL-C = low density lipoprotein cholesterol; MR = Mendelian randomization; NCBI = National Center for Biotechnology Information; PCa = prostate cancer; SHBG = sex hormone-binding globulin; SMR = summary data-based Mendelian randomization analysis; TC = total cholesterol; TG = triglycerides; TT = total testosterone; UKB = UK Biobank.

Figure 2.

MR analysis of association between HMGCR inhibition and 5 sex hormones/2 male diseases. (A) MR results of HMGCR inhibitors and 5 sex hormones; (B) MR results of HMGCR inhibitor and 2 male diseases (CHD was used to validate the effectiveness of HMGCR inhibitor). For continuous outcomes, we use β values (effect size) and 95% CI to represent the corresponding changes in outcome associated with a 1 SD decrease in LDL-C levels. For categorical outcomes, we calculate OR and 95% CI to indicate the event risk associated with a 1 SD decrease in LDL-C levels.

Figure 3.

Leave-one-out analysis of genetically proxied HMGCR inhibition on outcomes. (A) HMGCR inhibitor and total testosterone levels; (B) HMGCR inhibitor and bioavailable testosterone levels; (C) HMGCR inhibitor and sex hormone-binding globulin levels; (D) HMGCR inhibitor and estradiol levels. Leave-one-out analysis suggests that no SNP has a significant impact on the results (all rows are on the right side of 0).

Figure 4.

Leave-one-out analysis of genetically proxied HMGCR inhibition on outcomes. (A) HMGCR inhibitor and dehydroepiandrosterone levels; (B) HMGCR inhibitor and erectile dysfunction risk; (C) HMGCR inhibitor and prostate cancer risk; (D) HMGCR inhibitor and coronary heart disease risk. Leave-one-out analysis suggests that no SNP has a significant impact on the results (all rows are on the right side of 0).

Figure 5.

Visualized results of significant common causal variation between HMGCR express and TT/BT. The figure displays the chromosome loci of TT, BT inhibitor genes, the size of r2, and causal variant information. The source of HMGCR expression is LDL-C, and the figure also shows the lead SNP with the highest linkage selection.