Stroke and Alzheimer's Disease: A Mendelian Randomization Study

Front Genet. 2020 Jul 14:11:581. doi: 10.3389/fgene.2020.00581. eCollection 2020.

Abstract

Stroke and Alzheimer's disease (AD) are common neurological diseases. Several exiting studies indicated that late onset-AD and ischemic stroke have shared genetic links. Different kinds of stroke have different mechanisms. However, it remains unclear whether there is a causal relationship between different types of strokes, including any stroke (AS), any ischemic stroke (AIS), large-artery atherosclerotic stroke (LAS), and cardio-embolic stroke (CES), and AD. Herein, we conducted several Mendelian randomization (MR) studies to explore genetically causal link of different kinds of strokes and AD. The results for inverse-variance weighted (IVW) meta-analysis (β = -0.039, OR = 0.9618, and P-value = 0.750) and weighted median regression (WMR) (β = -0.156, OR = 0.8556, and P-value = 0.274) demonstrated that AS is not causally associated with AD risk. The result of MR-Egger regression (β = -1.312, P-value = 0.098) and intercept term (P-value = 0.105) illustrated no pleiotropy in this MR study. According to the results for IVW (P-value = 0.305, β = -0.103, and OR = 0.9021) and WMR (P-value = 0.487, β = -0.092, and OR = 0.9121) in the MR study between AIS and AD, there is no causal association between AIS and AD risk. In addition, the MR-Egger regression (P-value = 0.290 and β = -0.512) and intercept term (P-value = 0.387) showed no potential pleiotropy. LAS is not causally associated with AD risk according to the MR results (IVW: P-value = 0.568, β = 0.037, and OR = 1.0377; WMR: P-value = 0.793, β = -0.022, and OR = 0.9782). Additionally, the results of MR-Egger regression (P-value = 0.122 and β = -1.220) and intercept term (P-value = 0.110) showed no potential pleiotropy. Our results [IVW: P-value = 0.245, β = -0.064, and OR = 0.938; WMR: P-value = 0.331, β = -0.057, and OR = 0.9446; MR-Egger: P-value = 0.673 and β = -0.062, and intercept term (P-value = 0.985)] further demonstrated there is no causal link between CES and AD and no pleiotropy in this MR study. In conclusion, different types of stroke, including AS, AIS, LAS, and CES, would not be causally associated with AD risk.

Keywords: Alzheimer’s disease; Mendelian randomization; different kinds of stroke; genome-wide association study; pleiotropy.