Causal effects of genetically predicted testosterone on Alzheimer's disease: a two-sample mendelian randomization study.

Abstract

Although several studies have reported that testosterone may protect against Alzheimer's disease, no evidence of a causal relationship has been demonstrated. A Mendelian randomization (MR) study was performed to determine the causal role of testosterone in Alzheimer's disease. The study utilized public databases obtained from separately published genome-wide associationstudies (GWAS). Single-nucleotide polymorphisms (SNPs) for testosterone were extracted from the most recent and largest published GWAS meta-analysis (178,782 participants), and SNPs for Alzheimer's disease were extracted from UK Biobank (954 AD cases and 487,331 controls). The odds ratio (OR) of the inverse variance weighting (IVW) approach was the primary outcome, and the weighted median and MR Egger regression were used for sensitivity analysis. Through IVW, we observed a causal association between genetically predicted testosterone and the risk of Alzheimer's disease, with an OR of 0.99 (95% confidence interval [CI] = 0.998-0.999, p = 0.047). In the sensitivity analyses, the weighted median regression showed directionally similar estimates (OR = 0.99, 95% CI = 0.998-0.999, p = 0.048). The MR Egger regression showed similar estimates (OR = 0.99, 95% CI = 0.998-1.00, p = 0.35), but with lower precision. Funnel plots, MR Egger intercepts, and Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) analysis indicated the absence of directional pleiotropy effects. In conclusion, our MR study provides evidence of a causal relationship between testosterone levels and Alzheimer's disease; however, this relationship must be validated in future studies with larger sample sizes. Early testosterone monitoring may enable the prevention of Alzheimer's and related diseases.