Front. Genet. Frontiers in Genetics Front. Genet. 1664-8021 Frontiers Media S.A. 10.3389/fgene.2021.683082 Genetics Review Mendelian Randomization in Stroke: A Powerful Approach to Causal Inference and Drug Target Validation Acosta Julián N. 1 Szejko Natalia 1 2 3 Falcone Guido J. 1 * 1Division of Neurocritical Care and Emergency Neurology, Department of Neurology, Yale School of Medicine, New Haven, CT, United States 2Department of Neurology, Medical University of Warsaw, Warsaw, Poland 3Department of Bioethics, Medical University of Warsaw, Warsaw, Poland

Edited by: Lei Zhang, Soochow University, China

Reviewed by: Chaolong Wang, Huazhong University of Science and Technology, China; Lang Wu, University of Hawaii at Manoa, United States

 *Correspondence: Guido J. Falcone, guido.falcone@yale.edu

This article was submitted to Statistical Genetics and Methodology, a section of the journal Frontiers in Genetics

 12 08 2021 2021 12 683082 19 03 2021 28 06 2021 Copyright © 2021 Acosta, Szejko and Falcone. 2021 Acosta, Szejko and Falcone

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Stroke is a leading cause of death and disability worldwide. However, our understanding of its underlying biology and the number of available treatment options remain limited. Mendelian randomization (MR) offers a powerful approach to identify novel biological pathways and therapeutic targets for this disease. Around ~100 MR studies have been conducted so far to explore, confirm, and quantify causal relationships between several exposures and risk of stroke. In this review, we summarize the current evidence arising from these studies, including those investigating ischemic stroke, hemorrhagic stroke, or both. We highlight the different types of exposures that are currently under study, ranging from well-known cardiovascular risk factors to less established inflammation-related mechanisms. Finally, we provide an overview of future avenues of research and novel approaches, including drug target validation MR, which is poised to have a substantial impact on drug development and drug repurposing.

 stroke genetics Mendelian randomization polygenic risk scores drug target validation K76AG59992 R03NS112859 18IDDG34280056 817874 P30AG021342 National Institute on Aging10.13039/100000049 National Institute of Neurological Disorders and Stroke10.13039/100000065 American Heart Association10.13039/100000968 Yale Pepper Scholar Award Neurocritical Care Society Research Fellowship Introduction

Stroke is a leading cause of mortality and disability worldwide. While the overall incidence of stroke is decreasing, the global burden of this disease remains high because the absolute number of disability-adjusted life years lost due to stroke is increasing as the population grows and ages (Johnson et al., 2019). Thus, there is an urgent need for novel preventive, therapeutic, and rehabilitation strategies. Among these three, new acute treatments are particularly needed, as the few alternatives available thus far include thrombolytic therapy and mechanical thrombectomy for ischemic stroke (Powers et al., 2019), targeted blood pressure reduction for intracerebral hemorrhage (ICH; Hemphill et al., 2015), and early securing of the aneurysm in subarachnoid hemorrhage (Connolly et al., 2012).

In this setting, the field of stroke research can greatly benefit from the tools that population genetics has to offer. Mendelian randomization (MR) is a statistical method aimed at determining and quantifying causal relationships between genetically determined exposures and outcomes of interest (Davies et al., 2018). Importantly, in contrast to randomized clinical trials (the most frequently used tool to evaluate causality), MR can be performed using already available open-access data from different sources, allowing the evaluation of larger numbers of possible mechanisms and accelerating the speed of the translational cycle.

In this review, we introduced the concept of MR and provide an overview of the existing MR studies focused on stroke. To identify these studies, we conducted a systematic search of the medical literature using PubMed and the keywords “Mendelian randomization stroke,” “Mendelian randomization intracerebral hemorrhage,” and “Mendelian randomization subarachnoid hemorrhage” in the MEDLINE database. We also explored some analytical possibilities beyond classical MR, including the use of multiomics data and drug target validation MR.

 Mendelian Randomization: A Primer

MR constitutes a special case of instrumental variable analysis, a widely used analytical framework for causal inference. When the MR assumptions are met, it is possible to identify and quantify causal relationships between exposures and outcomes of interest (Figure 1). MR is best explained using an example. For this section, we will use the link between cholesterol levels and systolic blood pressure as our starting point. In randomized clinical trials, investigators randomly assign study participants to an intervention or placebo to study causal relationships between the intervention and the outcome of interest. In our example, researchers would randomly assign participants to receive a cholesterol-lowering drug or placebo and then measure systolic blood pressure levels in both groups. In MR analysis, we would use genetic variants strongly associated with the exposure of interest as the instrument. In this case, we would choose genetic variants strongly associated with cholesterol levels. As these variants are randomly segregated before birth, one could then separate groups according to the number of risk alleles, and the resulting analyses would not be confounded by environmental exposures happening after birth. Thus, we can measure systolic blood pressure levels in those groups and implement the necessary comparisons.

Mendelian randomization (MR) as instrument variable analysis. MR is a type of instrumental variable analysis, similar to randomized clinical trials. Exposed individuals in MR are those carrying risk alleles for determined genetic variants known to associate with an exposure of interest.

The implementation of MR analyses requires important assumptions (Figure 2). First, the genetic variants used as instruments must be strongly associated with the exposure or risk factor of interest. This is generally easily accomplished, since these variants are often derived from large-scale genome-wide association studies (GWAS) of the exposure of interest. Second, there should be no confounders affecting the association between genetic variants and the outcome of interest. This assumption is not trivial, and, at the moment, there is no method to definitively confirm this assumption. Although not always performed, a practical way to tackle this problem is to test the genetic variants used as instruments against an array of different other potential covariates that could lead to bias. Third, the genetic variants used as instruments affect the outcome of interest only via the exposure of interest, with no other alternative pathways coming into play. These alternative pathways are part of what is called horizontal pleiotropy. There are several statistical methods to test and correct for horizontal pleiotropy, including the Mendelian randomization–Egger (MR–Egger) intercept, MR Pleiotropy RESidual Sum and Outlier (MR-PRESSO), and Causal Analysis Using Summary Effect Estimates (CAUSE; Bowden et al., 2015; Verbanck et al., 2018; Morrison et al., 2020).

Mendelian randomization paradigm and assumptions. Mendelian randomization assumptions: (1) valid instrument, (2) no association with confounders, and (3) no horizontal pleiotropy.

Crucial to MR studies is the online availability of open-access resources, including tools to rapidly test hypotheses such as MR-Base (Hemani et al., 2018); large biobanks such as the UK Biobank (Bycroft et al., 2018), the China Kadoorie Biobank (CKB; Chen et al., 2011), Biobank Japan (Nagai et al., 2017), and the All of Us Research Program (Denny et al., 2019); and repositories of summary statistics from large GWAS like the GWAS catalog (Buniello et al., 2019), the GWAS atlas (Tian et al., 2020), and specifically for stroke, the Cerebrovascular Disease Knowledge Portal (Crawford et al., 2018). These and other helpful resources are summarized in Table 1.

Online resources available for Mendelian Randomization studies.

References Name Website Description
Crawford et al., 2018 Cerebrovascular Disease Knowledge Portal http://www.cerebrovascularportal.org/ A platform that allows for searching, visualizing, and analyzing variations related to cerebrovascular disease.
Buniello et al., 2019 GWAS Catalog https://www.ebi.ac.uk/gwas/ Catalog of GWAS results
Tian et al., 2020 GWAS atlas https://atlas.ctglab.nl/ Catalog of GWAS results
Lambert et al., 2019 PGS Catalog http://www.pgscatalog.org/ Catalog of Polygenic Risk Scores
Hemani et al., 2018 MR-base https://www.mrbase.org/ A database and analytical platform for Mendelian Randomization
Bycroft et al., 2018 UK Biobank https://www.ukbiobank.ac.uk/ Observational study, enrolling >500,000 participants, open-access
Gaziano et al., 2016 Million Veteran Program https://www.research.va.gov/mvp/ Observational study, enrolling 1 million participants, restricted access
Denny et al., 2019 All of Us https://allofus.nih.gov/ Observational study, aiming to enroll 1 million participants, open-access
Chen et al., 2011 China Kadoorie Biobank https://www.ckbiobank.org/ Observational study, 500,000 participants, restricted access
Nagai et al., 2017 Biobank Japan https://biobankjp.org/english/index.html Observational study, ~200,000 participants, restricted access
Wong et al., 2017 dbGAP https://www.ncbi.nlm.nih.gov/gap/ Repository, open-access
Lappalainen et al., 2015 EGA https://ega-archive.org/ Repository, open-access
Traditional Cardiovascular Risk Factors

Cardiovascular risk factors represent the most important determinants of risk of stroke. While several observational studies have shown associations between traditional vascular risk factors and risk of different types of stroke, deriving causality from observational data is problematic due to the possibility of bias. In the last 5 years, several MR studies have been conducted to explore, confirm, and quantify causality behind these already described associations (Table 2).

Summary of MR studies looking at traditional risk factors for stroke.

References Exposure Outcomes Findings
Georgakis et al., 2020b Blood pressure levels Any stroke, ischemic stroke, LAS, CES, and SVS Ten mmHg increase in SBP was associated with approximately 40% increase in the risk of any stroke or ischemic stroke. It was also associated with the risk of stroke subtypes, except for lobar ICH. Similar findings were presented for DBP. Decreases in SBP through calcium channel blockers but not through beta-blockers was associated with decrease in ischemic stroke risk.
Georgakis et al., 2020a PP Ischemic stroke and subtypes Pulse pressure was independently associated with stroke risk in participants older than 55 years.
Hindy et al., 2018 LDL-C Ischemic stroke and subtypes Higher LDL-C was associated with higher risk of ischemic stroke and LAS.
Allara et al., 2019 Lipid fractions levels CVD outcomes Higher LDL-C was associated with higher risk of ischemic stroke.Higher HDL-C levels were associated with lower risk of ICH.Higher triglycerides were associated with lower risk of ICH.
Valdes-Marquez et al., 2019 LDL-C CHD, Ischemic stroke and its subtypes LDL-C levels were not associated with the risk of ischemic stroke or its subtypes.
Georgakis et al., 2020c Lipid fractions levels SVS, WMH volume, and ICH Higher HDL-C levels were associated with lower risk of SVS and lower WMH volumes.Higher HDL-C levels were associated with higher risk of ICH.
Sun et al., 2019 Lipid fractions levels Ischemic stroke and ICH Higher LDL-C levels were associated with higher risk of ischemic stroke and lower risk of ICH.
Falcone et al., 2020 Lipid fractions levels ICH Higher LDL-C levels were associated with lower risk of ICH.
Pan et al., 2019 Lp(a) Ischemic stroke, its subtypes, AD Lp(a) levels were associated with higher risk of LAS, but lower risk of SVS and Alzheimer’s disease.
Larsson et al., 2020b Lp(a) Ischemic stroke, AD, parental AD or dementia Lp(a) levels were associated with the risk of ischemic stroke.There is an inverse relationship between Lp(a) and Parental AD or dementia.
Larsson et al., 2017 T2DM Ischemic stroke and its subtypes T2DM was associated with the risk of ischemic stroke, specially LAS.
Gan et al., 2019 T2DM Ischemic stroke T2DM was associated with ischemic stroke.
Liu et al., 2018 T2DM Small vessel disease phenotypes T2DM was associated with MRI-confirmed lacunar stroke.
Chen et al., 2020 Insulin resistance Ischemic stroke and subtypes Insulin resistance was associated with ischemic stroke, particularly SVS.
Yeung et al., 2018 HbA1c Ischemic stroke No association.
Georgakis et al., 2021 Type 2 diabetes, HbA1c, insulin resistance, and β-cell dysfunction Ischemic stroke, ischemic stroke subtypes, intracerebral hemorrhage, related phenotypes T2DM and higher HbA1c levels were associated with higher risk of ischemic stroke, especially LAS and SVS. Insulin resistance and β-cell dysfunction show similar associations, with the latter also associated with intracerebral hemorrhage. T2DM was also associated with lower white matter integrity (fractional anisotropy). T2DM, HbA1c, and β-cell dysfunction were associated with lower grey matter volume and total brain volume.
Qian et al., 2019 Smoking Ischemic stroke Smoking was associated with any ischemic stroke and LAS.
Larsson et al., 2019a Smoking initiation Ischemic stroke, subtypes, and ICH Smoking initiation was associated with ischemic stroke, LAS, and SVS, but not with CES or ICH.
Larsson et al., 2020c Smoking 14 CVDs Smoking was associated with a broad range of CVDs including coronary artery disease, heart failure, abdominal aortic aneurysm, ischemic stroke, transient ischemic attack, peripheral arterial disease, and arterial hypertension.
Acosta et al., 2021 Smoking initiation SAH Smoking initiation was associated with the risk of nontraumatic SAH.
Dale et al., 2017 General adiposity and central adiposity Stroke Central adiposity but not general adiposity was associated with stroke risk.
Marini et al., 2020 BMI and WHR Ischemic stroke, subtypes, ICH, WMH volume Higher WHR but not higher BMI was associated with all-cause ischemic stroke, LAS, SVS, non-lobar ICH and WMH volume.
Zhuang et al., 2020 Physical activity Stroke No association.
Hou et al., 2020 Atrial fibrillation CES Bidirectional association between atrial fibrillation and CES.

CVD, cardiovascular disease; ICH, intracerebral hemorrhage; LAS, large artery stroke; SVS, small vessel stroke; CES, cardioembolic stroke; WMH, white matter hyperintensity; SAH, subarachnoid hemorrhage; PP, pulse pressure; HDL-C, high-density lipoproteins cholesterol; LDL-C, low-density lipoprotein cholesterol; Lp(a), lipoprotein(a); T2DM, type 2 diabetes mellitus; HbA1c, glycosylated hemoglobin; BMI, body mass index; and WHR, waist-to-hip ratio.

 Blood Pressure

Robust existing evidence indicates that high blood pressure is one of the main risk factors for stroke. However, several questions remain, including how early and sustained changes in blood pressure affect outcomes later in life and whether specific pharmacological interventions provide additional benefits over others. Georgakis et al. (2020b) utilized MR analyses to assess the relationship between genetically proxied blood pressure levels and the risk of any stroke as well as its subtypes. They demonstrated that blood pressure levels were causally associated with the risk of any stroke and most subtypes, with the exception of lobar ICH. The effect of blood pressure appeared stronger in large artery stroke (LAS) and small vessel stroke (SVS). Additionally, the authors explored genetic proxies for two different antihypertensive drugs classes, calcium channel blockers and beta-blockers, finding that a 10 mmHg reduction in systolic blood pressure using genetic proxies for calcium channel blockers was protective for any stroke and its subtypes, especially for SVS (40% reduction). Furthermore, the same reduction in blood pressure through calcium channel blockers variants was also associated with lower white matter hyperintensity (WMH) volumes, a well-established neuroimaging biomarker for cerebral small vessel disease. A study from the same group also evaluated whether blood pressure pulsatility (i.e., pulse pressure) affects stroke risk independently of the mean arterial pressure. Using multivariable MR within the UK Biobank, they found that pulse pressure was independently associated with ischemic stroke risk (but not ICH) in participants older than 55 years, being particularly important for LAS (Georgakis et al., 2020a). Beyond ischemic stroke, MR analyses have confirmed observational evidence that indicate that high blood pressure is associated with higher risks of intracranial aneurysms and subarachnoid hemorrhage (Bakker et al., 2020).

 Lipids

The role of lipids in cardiovascular disease (CVD) has been extensively studied. However, the effect of circulating lipid levels on stroke appear to be heterogeneous and depend on the specific subtypes being evaluated. Hindy et al. (2018) investigated the effect of blood lipids on the risk of ischemic stroke and its subtypes using summary statistics from the GWAS completed by the Stroke Genetics Network (SiGN), which included 17,000 ischemic stroke cases and more than 32,000 controls of European ancestry. They found that low-density lipoprotein cholesterol (LDL-C) levels were associated with risk of ischemic stroke, especially LAS. HDL-C levels were inversely associated with the risk of SVS. There were no consistent associations for triglycerides. More recently, Allara et al. (2019) conducted a two-sample MR study to identify relationships between lipid levels and the risk of several cardiovascular outcomes, including stroke. In their study, higher HDL-C levels were associated with lower risk of ICH, while higher LDL-C levels were associated with higher risk of ischemic cerebrovascular disease. Interestingly, they also found an inverse association between triglycerides and the risk of ICH. A rigorous study performed by Valdes-Marquez et al. (2019) investigated the effect of LDL-C on the risk of ischemic stroke and its subtypes and did not find any significant associations with these conditions. Georgakis et al. (2020c) also looked at the association between lipid fractions and small vessel disease, a broad phenotype that includes SVS, white matter disease, and ICH. They found that higher HDL-C was associated with lower risk of SVS and lower WMH volume. Interestingly, they also found a relationship between higher HDL-C and higher risk of ICH. Along this lines, Sun et al. (2019) found a strong positive association between LDL-C and ischemic stroke and a strong inverse association with ICH in participants of the China Kadoorie Biobank. This inverse association between genetically determined LDL-C levels and the risk of ICH was confirmed in cohorts of European descent by Falcone et al. (2020). The latter two studies confirmed previous observational results that suggested that very low LDL-C levels increase the risk of ICH (Wang et al., 2013; Saliba et al., 2018; Ma et al., 2019), pointing to novel pathways in ICH.

Another study recently explored association pertains to lipoprotein(a) [Lp(a)] levels and ischemic stroke. Using summary statistics from MEGASTROKE and the International Genomics of Alzheimer’s Project (IGAP), Pan et al. (2019) found a causal positive association between Lp(a) levels and LAS and, remarkably, an inverse association with SVS and Alzheimer disease (AD). Using individual-level data from the UK Biobank, Larsson et al. (2020b) confirmed a positive association with any ischemic stroke but could not replicate the protective association with AD. The latter result, however, could have been due to lack of statistical power, as the investigators did find a protective association with parental AD or dementia.

 Type 2 DM

Extensive data point to a deleterious effect of diabetes on ischemic stroke, but there is no such evidence for hemorrhagic stroke. Larsson et al. (2017) used MR analyses to confirm a causal relationship between type 2 diabetes mellitus (T2DM) and risk of ischemic stroke, especially LAS, but found null associations between other metabolic markers such as glucose, insulin levels, and body mass index (BMI). Gan et al. (2019) replicated this causal association between T2DM and ischemic stroke in the CKB, and Liu et al. (2018) found an association between T2DM and lacunar stroke. Similarly, using a combined exposure comprising phenotypes related to insulin resistance (fasting insulin adjusted for BMI, HLD-C and triglycerides, and insulin sensitivity), Chen et al. (2020) found a causal relationship between insulin resistance and ischemic stroke, particularly small vessel stroke. On the other hand, Yeung et al. (2018) did not find an association between glycated hemoglobin (HbA1c) and ischemic stroke in an MR study within the UK Biobank. More recently, Georgakis et al. (2021) investigated the effects of T2DM, hyperglycemia, insulin resistance, and β-cell dysfunction on the risk of stroke and related traits, finding that type 2 diabetes and higher HbA1c levels are associated with higher risk stroke, and particularly of LAS and SVS, with similar associations found for insulin resistance and β-cell dysfunction. β-Cell dysfunction was also associated with the risk of ICH. Furthermore, a study focused on neuroimaging markers of cerebrovascular disease found that type 2 diabetes was also positively associated with fractional anisotropy (a measure of white matter integrity) and inversely associated with gray matter volume and total brain volume, with similar inverse associations seen for HbA1c and β-cell dysfunction.

 Smoking

A large body of literature indicate that smoking is associated with both ischemic stroke and aneurysmal subarachnoid hemorrhage (Peters et al., 2013). Qian et al. (2019) reported a causal association between smoking and any ischemic stroke and LAS. In other MR studies, Larsson et al. (2019a, 2020c) confirmed the association between smoking initiation and the risk of any ischemic stroke, LAS and SVS, but did not find an association with cardioembolic stroke (CES) or ICH. In a one-sample MR study using the UK Biobank, Acosta et al. (2021) confirmed an association between genetic propensity to smoke and risk of nontraumatic subarachnoid hemorrhage. Similarly, in a large GWAS of intracranial aneurysms and subarachnoid hemorrhage, smoking was shown to be a powerful risk factor for this disease (Bakker et al., 2020).

 Obesity

Observational evidence suggests that obesity is a risk factor for stroke. However, MR studies have yielded heterogeneous results. One study indicated that while both general and central adiposity had causal effects on coronary heart disease, only central adiposity appeared to be associated with ischemic stroke (Dale et al., 2017). Further evidence supporting this hypothesis was provided by a more recent MR study by Marini et al. (2020) which found that higher waist-to-hip ratio, but not higher BMI, was causally associated with all-cause ischemic stroke, LAS, SVS, non-lobar ICH, and WMH volume.

 Lack of Physical Activity

Physical activity, especially moderate-to-vigorous physical activity, has been associated with lower risk of stroke in several observational studies. To date, only one MR has investigated this relationship, finding null results for this association (Zhuang et al., 2020). Therefore, further research is warranted in order to fully understand these conflicting results.

 Atrial Fibrillation

Atrial fibrillation (AF) is considered a major risk factor of cardioembolic ischemic stroke. Remarkably, patients with stroke are also at a higher risk of developing AF (Sposato et al., 2015, 2018). Only one MR study has looked at this relationship and, as expected given the overwhelming amount of observational evidence, found causal evidence to support this link using bidirectional MR (Hou et al., 2020).

 Nontraditional Risk Factors

A number of studies have explored nontraditional risk factors that had mixed or inconclusive evidence when evaluated in nongenetic, observational studies. Prominent among these are studies that focused on inflammation, coagulation factors, sleep health, and nutrition. MR studies have evaluated several of these factors in search of causal associations (Table 3).

Summary of MR studies looking at nontraditional risk factors for stroke.

References Exposure Outcomes Findings
Larsson et al., 2020a Alcohol CVDs Causal relationship between higher alcohol consumption and increased risk of stroke and peripheral artery disease.
Yuan et al., 2020c PUFA IS High-level plasma ALA was protective for IS, but AA was the opposite. LA, EPA, DHA, and DPA had no effects on IS.
Au Yeung and Schooling, 2020 Urinary sodium CVDs Higher log-transformed urinary sodium was associated with higher risk of stroke.
Larsson et al., 2019b Serum magnesium and calcium IS Genetically higher serum magnesium concentrations were associated with a reduced risk of cardioembolic stroke but found no significant association of genetically higher serum calcium concentrations with any IS subtype.
Wang et al., 2021 Tea IS Genetically predicted an extra daily cup of tea consumption was causally associated with a reduced risk of small vessel stroke.
Qian et al., 2020 Coffee IS and ICH Coffee consumption was not causally associated with risk of stroke or its subtypes.
Choi et al., 2020 Serum bilirubin levels IS an ICH Causal associations between serum bilirubin levels and decreased stroke risk.
Huang et al., 2019 25(OH)D CVDs No evidence to support that genetically increased 25(OH)D was associated with a lower risk of IS, ICH, and SAH.
Larsson et al., 2018b 25(OH)D IS No evidence that genetically determined higher S-25OHD concentrations were causally associated with any ischemic stroke subtype.
Larsson et al., 2018a VitK1 IS Genetic predisposition to higher circulating vitamin K1 levels was associated with an increased risk of large artery atherosclerotic stroke.
Schooling et al., 2018 Testosterone CVDs Genetically determined testosterone was associated with IS.
Zhang et al., 2020 CRP IS No clear support that genetically determined elevated CRP concentration was causally associated with the risk of IS.
Yuan et al., 2020b Circulating ILs CVDs Positive association of IL-1 with cardioembolic stroke and suggestive inverse associations of IL-6 with any IS, cardioembolic stroke, and small vessel stroke, and of IL-16 with CAD.
Georgakis et al., 2020d Genetic proxies for IL-6R-mediated downregulation of IL-6 signaling IS and other CVDs Genetically downregulated IL-6 signaling to be associated with lower risks of IS.
Yuan et al., 2020a TNF CVDs Genetically predicted TNF levels were positively associated with coronary artery disease and IS.
Wang et al., 2020 GDF-15 CVDs Causal relationship of circulating GDF-15 levels with the increased risk of cardioembolic stroke, atrial fibrillation, coronary artery disease and myocardial infarction, but not any IS, large-artery atherosclerotic stroke, small vessel stroke, heart failure, and nonischemic cardiomyopathy.
Song et al., 2020 TIM-1 Stroke Causal effect of TIM-1 on stroke.
Georgakis et al., 2019 Cytokines Stroke Genetic predisposition to elevated circulating levels of MCP-1 was associated with higher risk of stroke, in particular with large-artery stroke and cardioembolic stroke.
Lin et al., 2020 Inflammatory biomarkers IS No convincing evidence to support that inflammatory biomarkers like IL-1Ra, sIL-6R, and CRP were causally associated with the risk of IS or its subtypes.
Gill et al., 2018c Iron Stroke MR evidence that higher iron status was associated with increased stroke risk and, in particular, CES.
Titova et al., 2020 Sleep duration Stroke No clear support that a genetically determined short or long sleep duration has influence on the risk of total stroke or stroke types.
Lu et al., 2020 Sleep duration Stroke Sleep duration was not causally associated with risk of stroke and its subtypes.
Cai et al., 2020 Sleep traits IS Potential causal role of short sleep duration and insomnia symptoms in LAS.
Larsson and Markus, 2019 Insomnia CAD and stroke Causal link between insomnia and ischemic stroke and its subtypes.
Marouli et al., 2020 Thyroid hormones Stroke or CAD A 1-SD increase in TSH was associated with a 5% decrease in the risk of stroke.
Cai et al., 2019 MDD SVS Possible causal effect of MDD on increased risk of SVS.
Gill et al., 2019b Depression IS and functional outcome after IS No evidence of genetically determined risk of depression affecting IS risk but consistent MR evidence suggestive of a possible effect on functional outcome after IS.
Gill et al., 2018b Platelet count CVDs Higher genetically determined platelet count was causally associated with higher risk of IS.
Gill et al., 2018a FXI IS, ICH, MI Causal effect of higher, genetically determined FXI levels on risk of any IS.
Harshfield et al., 2020 Hematological traits IS and its subtypes Several factors on the intrinsic clotting pathway were significantly associated with CES and LAS, but not with SVS. On the common pathway, increased gamma (γ’) fibrinogen was significantly associated with AIS/CES. Furthermore, elevated plateletcrit was significantly associated with AIS/CES, eosinophil percentage of white cells with LAS, and thrombin-activatable fibrinolysis inhibitor activation peptide antigen with AIS. Follow-up analysis in UK Biobank showed that among individuals with atrial fibrillation, those with genetically lower levels of factor XI are at reduced risk of AIS compared to those with normal levels of factor XI.
Jia et al., 2019 Gut microbiota dependent metabolites (betaine, carnitine, choline, and trimethylamine N-oxide) Stroke No association.

CVDs, cardiovascular disorders; ICH, intracerebral hemorrhage; IS, ischemic stroke; PUFA, polyunsaturated fatty acids; AA, arachidonic acid; LA, linoleic acid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; DPA, docosapentaenoic acid; SAH, subarachnoid hemorrhage; FAs, circulating fatty acids; CRP, C-reactive protein; ILs, interleukins; IL-1, interleukin 1; IL-6, interleukin 6; IL-16, interleukin 16; TNF, tumor necrosis factor; GDF-15, growth differentiation factor 15; TIM-1, T-cell immunoglobulin and mucin-1; MCP-1, monocyte chemoattractant protein-1; LAS, large artery stroke; CAD, coronary artery disease; TSH, thyroid stimulating hormone; MDD, major depressive disorder; SVS, small vessel stroke; MR, Mendelian randomization; FXI, factor XI; MI, myocardial infarction; CES, cardioembolic stroke; and AIS, arterial ischemic stroke.

 Inflammatory Biomarkers

Observational studies have demonstrated that patients with underlying inflammation have a higher risk of stroke (Anrather and Iadecola, 2016), although this association could be the result of confounding. Therefore, genetically determined inflammatory biomarkers have been targeted by many studies investigating the risk of stroke. Zhang et al. (2020) examined whether genetically raised plasma C-reactive protein (CRP) concentration levels were associated with ischemic stroke without finding a significant association. In line with these findings, Lin et al. (2020) evaluated numerous inflammatory biomarkers and found no association between genetically elevated levels of these biomarkers and ischemic stroke. Yuan et al. (2020b) analyzed genetically determined circulating interleukins in relation to coronary artery disease (CAD), atrial fibrillation, and ischemic stroke, and its subtypes. There was a suggestive (p < 0.05 but not statistically significant after correction for multiple testing) positive association between interleukin-1 receptor antagonist and cardioembolic stroke and a suggestive inverse association between interleukin-6 and ischemic stroke, CES, and SVS, and of interleukin-16 with CAD. Georgakis et al. (2020d) reported the results of an MR study focused on interleukin-6 signaling effects on ischemic stroke and other cardiovascular outcomes and demonstrated that genetically downregulated interleukin-6 signaling was associated with lower risk of ischemic stroke. The same group published a study investigating genetically determined levels of circulating cytokines and risk of stroke (Georgakis et al., 2019). They showed that genetic predisposition to elevated circulating levels of monocyte chemoattractant protein-1 was associated with higher risk of stroke, in particular with LAS and CES. Another study by Yuan et al. (2020a) investigated causal associations of increased tumor necrosis factor levels and several highly prevalent CVDs. Genetically elevated tumor necrosis factor levels were positively associated with both CAD and ischemic stroke. In addition, Wang et al. (2020) investigated the impact of growth differentiation factor 15 on the risk of CVDs using an MR approach. They found evidence for a relationship between circulating of growth differentiation factor 15 levels and increased risk of CES, atrial fibrillation, CAD, and myocardial infarction. Finally, Song et al. (2020) published results of their study investigating association of genetically determined T-cell immunoglobulin and mucin domain 1 with incidence of stroke, which found a causal effect of TIM-1 on any stroke and ischemic stroke.

 Hematological Traits

The role of hematological traits and pathways in the occurrence of stroke has also been extensively evaluated using MR. Gill et al. (2018b) investigated genetically determined platelet count and risk of different CVDs, finding that higher genetically determined platelet count is associated with higher risk of ischemic stroke. The same group released results of an MR study focused on genetically determined factor XI levels that found causal evidence supporting factor XI as a possible target to reduce the risk of cardioembolic stroke (Gill et al., 2018a). Harshfield et al. (2020) reported a study that, rather than focusing on a single target, evaluated several different hematological traits in connection to risk of ischemic stroke and its subtypes. Several factors in the intrinsic coagulation pathway were significantly associated with CES and LAS but not with SVS. Specifically, gamma fibrinogen, a component of the common pathway, was associated with CES, plateletcrit was associated with CES, eosinophil percentage of white cells was associated with LAS, and thrombin-activatable fibrinolysis inhibitor activation peptide antigen was associated with any ischemic stroke. Follow-up analyses in the UK Biobank showed that among individuals with atrial fibrillation, those with genetically lower versus normal levels of factor XI have a reduced risk of ischemic stroke. Finally, one group investigated circulating vitamin K1 levels in connection to cerebrovascular disease and found that genetic predisposition to higher circulating vitamin K₁ levels was associated with an increased risk of LAS (Larsson et al., 2018a).

 Nutritional Factors

A number of MR studies also investigated the role of nutritional factors in the occurrence of stroke, another research avenue extensively explored from an observational perspective. Larsson et al. (2020a) examined whether genetically determined predisposition to alcohol consumption had influence on risk of CVD. This study provided evidence for a causal relationship between higher alcohol consumption and increased risk of any stroke and peripheral artery disease.

Beyond alcohol, several other nutritional factors have been investigated. One research group examined the role of plasma phospholipid fatty acids in risk of 15 CVD-related phenotypes. Genetically higher plasma α-linolenic, linoleic, and oleic acid levels were inversely associated with LAS and venous thromboembolism, whereas arachidonic and stearic acid levels are positively associated with these two CVD-related outcomess (Yuan et al., 2019).

Iron metabolism, including anemia and polycythemia, have long been postulated to play a role in cerebrovascular disease. Gill et al. (2018c) published results of an MR study investigating this question, finding that higher iron levels were associated with increased risk of stroke and, in particular, CES.

Two groups reported on genetically predicted levels of vitamin D and the risk of stroke. Huang et al. investigated whether vitamin D played a role in risk and mortality of several vascular diseases by conducting an MR study that included both Asian and European participants (Huang et al., 2019). They found no evidence to support that genetically increased vitamin D was associated with a lower risk of ischemic stroke, ICH, SAH, and lipid levels in neither Chinese nor European. Similarly, Larsson et al. (2018b) reported the results of a study on genetically determined vitamin D concentrations and ischemic stroke and its subtypes that failed to find significant associations.

 Other Risk Factors

Among other investigated risk factors, it is worth mentioning two studies investigating endocrine changes. Schooling et al. (2018) analyzed genetic predictors of testosterone and their associations with different CVD phenotypes. They confirmed prior results from observational studies showing a significant association between genetically proxied testosterone and risk of ischemic stroke. Another study by Marouli et al. (2020) used MR analyses to analyze whether thyroid function affects the risk of stroke via atrial fibrillation, finding that a 1 SD increase in TSH was associated with a 5% decrease in the risk of ischemic stroke.

Finally, another four studies are also worth mentioning. Larsson et al. (2019b) examined serum magnesium and calcium levels in relation to ischemic stroke using MR. They found that genetically higher serum magnesium concentrations were associated with a reduced risk of CES but not with other stroke subtypes. Au Yeung and Schooling (2020) investigated the impact of urinary sodium on CVD. Higher genetically determined log-transformed urinary sodium was associated with higher risk of stroke. Choi et al. (2020) used MR tools to describe a causal association between serum bilirubin levels and decreased stroke risk. Lastly, Jia et al. (2019) investigated the role of gut microbiota-dependent metabolites and risk of several CVDs, finding no evidence of a causal link between these metabolites (betaine, carnitine, choline, and trimethylamine N-oxide) and the risk of stroke.

 Mendelian Randomization for Drug Target Validation and Drug Repurposing

Beyond hypothesis-driven MR studies, population genetics offers powerful tools to accelerate the discovery of novel biological pathways by agnostically evaluating several biological targets and/or reevaluating targets for drug repurposing (Nelson et al., 2015; Schmidt et al., 2020). This approach is significantly potentiated by the growing culture of open-access research and the increasing availability of high-throughput genomic and proteomic technologies.

Drug target MR studies use genetic variants that lie within or near the genes coding for these targets (the latter called cis-variants) as instruments, which either have an effect on the actual serum levels of the target protein or other endpoint, such as an intermediate biomarker, gene expression levels, or metabolite levels. This distinction is important because drug target MR aims to answer a different question than conventional MR. While conventional MR establishes causal relationships between biomarkers or traits and an outcome, drug target MR addresses whether modifications of a specific drug target or protein will have an effect on the outcome (Gill et al., 2021).

Drug target MR overcomes problems related to pleiotropy, which are especially relevant when looking at possible targets for intervention. In addition, by testing these drug targets phenome-wide, investigators can also pinpoint possible adverse effects. While drug target MR is a robust methodology, it also has limitations. While new high-throughput technologies have increased the amount of proteome-wide data available, publicly available summary statistics and proteomic studies with adequate sample sizes are still limited. This relative paucity of proteomic data leads to the utilization of cis-variants associated with a biomarker in the causal pathway, which could in turn lead to some problems. For example, circulating levels of a biomarker could not represent accurately its cellular concentration, which is often the value of interest, and this could limit the ability to detect causality. Additionally, the biomarker of interest could be relevant only in certain physiological or disease states, or during a critical period of time, which could also limit results. Lastly, not all proteins are druggable (i.e., able to be pharmacologically manipulated), which of course would defeat the purpose of doing drug target MR (Mokry et al., 2015).

Gill et al. (2019a) compared the results of drug target MR and clinical trials for three antihypertensive drug classes (angiotensin-converting-enzyme inhibitors, β-blockers, and calcium channel blockers) in coronary heart disease and stroke, finding comparable estimates. Along these lines, Chong et al. (2019) conducted a proteome-wide MR study to investigate potential therapeutic targets in ischemic stroke. The authors analyzed 653 circulating proteins as possible causal factors for the three main subtypes of ischemic strokes (LAS, SVS, and CES) and hemorrhagic stroke. In their analyses, they found eight biomarker–stroke associations encompassing seven unique targets. Of these biomarkers, five had already been associated with CVDs, including the coagulation factor 11, Lp(a), ABO, CD40 (a member of the tumor necrosis factor superfamily), and MMP12 (a member of the matrix metalloproteinase family, implicated in vascular remodeling). Novel biomarkers included SCARA5, a protein with a role in iron homeostasis, and TNFS12, a pleiotropic tumor necrosis factor-like cytokine linked to atrial fibrillation, a potential mediating mechanism.

Drug repurposing is another potentially useful approach, as previously approved drugs can be more easily brought to clinical practice if a beneficial effect is found. There are no clear examples of MR studies specifically focused on drug repurposing for stroke. However, we bring the reader’s attention to two such studies that explored the potential of repurposing medications to treat Alzheimer’s disease, unfortunately with null results (Walker et al., 2020; Wu et al., 2021). These studies provide an appropriate example and analytical framework for future studies applying this approach to stroke and cerebrovascular diseases.

 Conclusion

Stroke constitutes an increasingly prevalent condition worldwide and remains one of the leading causes of death and disability. MR has proved to be a powerful methodology to confirm or refute associations described by observational studies and identify novel therapeutic targets for stroke. The ability of MR studies to add valuable scientific evidence to the field of cerebrovascular disease research will be greatly increased by the utilization of novel tools, including proteome-wide and drug target validation analyses.

 Author Contributions

JA, NS, and GF reviewed the literature, drafted the manuscript, and revised the manuscript for intellectual content. All authors contributed to the article and approved the submitted version.

 Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

 Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

 References Acosta J. N. Szejko N. Both C. P. Vanent K. Noche R. B. (2021). Genetically determined smoking behavior and nontraumatic subarachnoid hemorrhage. Stroke 52, 582–587. doi: 10.1161/STROKEAHA.120.031622, PMID: 33440997 Allara E. Morani G. Carter P. Gkatzionis A. Zuber V. Foley C. N. . (2019). Genetic determinants of lipids and cardiovascular disease outcomes: a wide-angled Mendelian randomization investigation. Circ. Genom. Precis. Med. 12, 543–551. doi: 10.1161/CIRCGEN.119.002711 Anrather J. Iadecola C. (2016). Inflammation and stroke: an overview. Neurotherapeutics 13, 661–670. doi: 10.1007/s13311-016-0483-x, PMID: 27730544 Au Yeung S. L. Schooling C. M. (2020). Impact of urinary sodium on cardiovascular disease and risk factors: a 2 sample Mendelian randomization study. Clin. Nutr. 40, 1990–1996. doi: 10.1016/j.clnu.2020.09.018, PMID: 33004230 Bakker M. K. van der Spek R. A. A. van Rheenen W. Morel S. Bourcier R. Hostettler I. C. . (2020). Genome-wide association study of intracranial aneurysms identifies 17 risk loci and genetic overlap with clinical risk factors. Nat. Genet. 52, 1303–1313. doi: 10.1038/s41588-020-00725-7, PMID: 33199917 Bowden J. Smith G. D. Burgess S. (2015). Mendelian randomization with invalid instruments: effect estimation and bias detection through egger regression. Int. J. Epidemiol. 44, 512–525. doi: 10.1093/ije/dyv080, PMID: 26050253 Buniello A. MacArthur J. A. L. Cerezo M. Harris L. W. Hayhurst J. Malangone C. . (2019). The NHGRI-EBI GWAS catalog of published genome-wide association studies, targeted arrays and summary statistics 2019. Nucleic Acids Res. 47, D1005–D1012. doi: 10.1093/nar/gky1120, PMID: 30445434 Bycroft C. Freeman C. Petkova D. Band G. Elliott L. T. Sharp K. . (2018). The UK biobank resource with deep phenotyping and genomic data. Nature 562, 203–209. doi: 10.1038/s41586-018-0579-z, PMID: 30305743 Cai H. Cai B. Zhang H. Sun W. Wang Y. Zhou S. . (2019). Major depression and small vessel stroke: a Mendelian randomization analysis. J. Neurol. 266, 2859–2866. doi: 10.1007/s00415-019-09511-w, PMID: 31435769 Cai H. Liang J. Liu Z. Fang L. Zheng J. Xu J. . (2020). Causal effects of sleep traits on ischemic stroke and its subtypes: a Mendelian randomization study. Nat. Sci. Sleep 12, 783–790. doi: 10.2147/NSS.S265946 Chen Z. Chen J. Collins R. Guo Y. Peto R. Wu F. . (2011). China Kadoorie biobank of 0.5 million people: survey methods, baseline characteristics and long-term follow-up. Int. J. Epidemiol. 40, 1652–1666. doi: 10.1093/ije/dyr120, PMID: 22158673 Chen W. Wang S. Lv W. Pan Y. (2020). Causal associations of insulin resistance with coronary artery disease and ischemic stroke: a Mendelian randomization analysis. BMJ Open Diabetes Res. Care 8:e001214. doi: 10.1136/bmjdrc-2020-001217 Choi Y. Lee S. J. Spiller W. Jung K. J. Lee J. Y. Kimm H. . (2020). Causal associations between serum bilirubin levels and decreased stroke risk a two-sample Mendelian randomization study. Arterioscler. Thromb. Vasc. Biol. 40, 437–445. doi: 10.1161/ATVBAHA.119.313055, PMID: 31801373 Chong M. Sjaarda J. Pigeyre M. Mohammadi-Shemirani P. Lali R. Shoamanesh A. . (2019). Novel drug targets for ischemic stroke identified through Mendelian randomization analysis of the blood proteome. Circulation 140, 819–830. doi: 10.1161/CIRCULATIONAHA.119.040180, PMID: 31208196 Connolly E. S. Rabinstein A. A. Carhuapoma J. R. Derdeyn C. P. Dion J. Higashida R. T. . (2012). Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American heart association/American stroke association. Stroke 43, 1711–1737. doi: 10.1161/STR.0b013e3182587839, PMID: 22556195 Crawford K. M. Gallego-Fabrega C. Kourkoulis C. Miyares L. Marini S. Flannick J. . (2018). Cerebrovascular disease knowledge portal an open-access data resource to accelerate genomic discoveries in stroke. Stroke 49, 470–475. doi: 10.1161/STROKEAHA.117.018922, PMID: 29335331 Dale C. E. Fatemifar G. Palmer T. M. White J. Prieto-Merino D. Zabaneh D. . (2017). Causal associations of adiposity and body fat distribution with coronary heart disease, stroke subtypes, and type 2 diabetes mellitus: a Mendelian randomization analysis. Circulation 135, 2373–2388. doi: 10.1161/CIRCULATIONAHA.116.026560, PMID: 28500271 Davies N. M. Holmes M. V. Smith D. G. (2018). Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ 362:k601. doi: 10.1136/bmj.k601, PMID: 30002074 Denny J. C. Rutter J. L. Goldstein D. B. Philippakis A. Smoller J. W. Jenkins G. . (2019). The “All of Us” research program. N. Engl. J. Med. 381, 668–676. doi: 10.1056/NEJMsr1809937, PMID: 31412182 Falcone G. J. Kirsch E. Acosta J. N. Noche R. B. Leasure A. Marini S. . (2020). Genetically elevated LDL associates with lower risk of intracerebral hemorrhage. Ann. Neurol. 88, 56–66. doi: 10.1002/ana.25740, PMID: 32277781 Gan W. Bragg F. Walters R. G. Millwood I. Y. Lin K. Chen Y. . (2019). Genetic predisposition to type 2 diabetes and risk of subclinical atherosclerosis and cardiovascular diseases among 160,000 Chinese adults. Diabetes 68, 2155–2164. doi: 10.2337/db19-0224, PMID: 31399431 Gaziano J. M. Concato J. Brophy M. Fiore L. Pyarajan S. Breeling J. . (2016). Million veteran program: a mega-biobank to study genetic influences on health and disease. J. Clin. Epidemiol. 70, 214–223. doi: 10.1016/j.jclinepi.2015.09.016, PMID: 26441289 Georgakis M. K. Gill D. Malik R. Protogerou A. D. Webb A. J. S. Dichgans M. (2020a). Genetically predicted blood pressure across the lifespan differential effects of mean and pulse pressure on stroke risk. Hypertension 76, 953–961. doi: 10.1161/HYPERTENSIONAHA.120.15136, PMID: 32623925 Georgakis M. K. Gill D. Rannikmäe K. Traylor M. Anderson C. D. Lee J. M. . (2019). Genetically determined levels of circulating cytokines and risk of stroke: role of monocyte chemoattractant protein-1. Circulation 139, 256–268. doi: 10.1161/CIRCULATIONAHA.118.035905, PMID: 30586705 Georgakis M. K. Gill D. Webb A. Evangelou E. Elliott P. Sudlow C. . (2020b). Genetically determined blood pressure, antihypertensive drug classes and risk of stroke subtypes. Neurology 95, e353–e361. doi: 10.1212/WNL.0000000000009814, PMID: 32611631 Georgakis M. K. Harshfield E. L. Malik R. Franceschini N. Langenberg C. Wareham N. J. . (2021). Diabetes mellitus, glycemic traits, and cerebrovascular disease: a Mendelian randomization study. Neurology 96, e1732–e1742. doi: 10.1212/WNL.0000000000011555, PMID: 33495378 Georgakis M. K. Malik R. Anderson C. D. Parhofer K. G. Hopewell J. C. Dichgans M. (2020c). Genetic determinants of blood lipids and cerebral small vessel disease: role of high-density lipoprotein cholesterol. Brain 143, 597–610. doi: 10.1093/brain/awz413 Georgakis M. K. Malik R. Gill D. Franceschini N. Sudlow C. L. M. Dichgans M. (2020d). Interleukin-6 signaling effects on ischemic stroke and other cardiovascular outcomes: a Mendelian randomization study. Circ. Genom. Precis. Med. 13:e002872. doi: 10.1161/CIRCGEN.119.002872 Gill D. Georgakis M. K. Koskeridis F. Jiang L. Feng Q. Wei W. Q. . (2019a). Use of genetic variants related to antihypertensive drugs to inform on efficacy and side effects. Circulation 140, 270–279. doi: 10.1161/CIRCULATIONAHA.118.038814 Gill D. Georgakis M. K. Laffan M. Sabater-Lleal M. Malik R. Tzoulaki I. . (2018a). Genetically determined FXI (factor XI) levels and risk of stroke. Stroke 49, 2761–2763. doi: 10.1161/STROKEAHA.118.022792 Gill D. Georgakis M. K. Walker V. M. Schmidt A. F. Gkatzionis A. Freitag D. F. . (2021). Mendelian randomization for studying the effects of perturbing drug targets. Wellcome Open Res. 6:16. doi: 10.12688/wellcomeopenres.16544.2, PMID: 33644404 Gill D. James N. E. Monori G. Lorentzen E. Fernandez-Cadenas I. Lemmens R. . (2019b). Genetically determined risk of depression and functional outcome after ischemic stroke: Mendelian randomization study. Stroke 50, 2219–2222. doi: 10.1161/STROKEAHA.119.026089 Gill D. Monori G. Georgakis M. K. Tzoulaki I. Laffan M. (2018b). Genetically determined platelet count and risk of cardiovascular disease: Mendelian randomization study. Arterioscler. Thromb. Vasc. Biol. 38, 2862–2869. doi: 10.1161/ATVBAHA.118.311804 Gill D. Monori G. Tzoulaki I. Dehghan A. (2018c). Iron status and risk of stroke: a Mendelian randomization study. Stroke 49, 2815–2821. doi: 10.1161/STROKEAHA.118.022701 Harshfield E. L. Sims M. C. Traylor M. Ouwehand W. H. Markus H. S. (2020). The role of haematological traits in risk of ischaemic stroke and its subtypes. Brain 143, 210–221. doi: 10.1093/brain/awz362, PMID: 31755939 Hemani G. Zheng J. Elsworth B. Wade K. H. Haberland V. Baird D. . (2018). The MR-base platform supports systematic causal inference across the human phenome. eLife 7, 1–29. doi: 10.7554/eLife.34408 Hemphill J. C. Greenberg S. M. Anderson C. S. Becker K. Bendok B. R. Cushman M. . (2015). Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American heart association/American stroke association. Stroke 46, 2032–2060. doi: 10.1161/STR.0000000000000069, PMID: 26022637 Hindy G. Engström G. Larsson S. C. Traylor M. Markus H. S. Melander O. . (2018). Role of blood lipids in the development of ischemic stroke and its subtypes: a Mendelian randomization study. Stroke 49, 820–827. doi: 10.1161/STROKEAHA.117.019653, PMID: 29535274 Hou L. Xu M. Yu Y. Sun X. Liu X. Liu L. . (2020). Exploring the causal pathway from ischemic stroke to atrial fibrillation: a network Mendelian randomization study. Mol. Med. 26, 1–9. doi: 10.1186/s10020-019-0133-y, PMID: 31941463 Huang T. Afzal S. Yu C. Guo Y. Bian Z. Yang L. . (2019). Vitamin D and cause-specific vascular disease and mortality: a Mendelian randomisation study involving 99,012 Chinese and 106,911 European adults. BMC Med. 17:160. doi: 10.1186/s12916-019-1401-y, PMID: 31466528 Jia J. Dou P. Gao M. Kong X. Li C. Liu Z. . (2019). Assessment of causal direction between gut microbiota- dependent metabolites and cardiometabolic health: a bidirectional Mendelian randomization analysis. Diabetes 68, 1747–1755. doi: 10.2337/db19-0153, PMID: 31167879 Johnson C. O. Nguyen M. Roth G. A. Nichols E. Alam T. Abate D. . (2019). Global, regional, and national burden of stroke, 1990–2016: a systematic analysis for the global burden of disease study 2016. Lancet Neurol. 18, 439–458. doi: 10.1016/S1474-4422(19)30034-1, PMID: 30871944 Lambert S. A. Jupp S. Abraham G. Parkinson H. Danesh J. MacArthur J. A. L. . (2019). The Polygenic Score (pgs) Catalog: A Database of Published PGS to Enable Reproducibility and Uniform Evaluation. Available at: https://www.pgscatalog.org (Accessed February 1, 2021). Lappalainen I. Almeida-King J. Kumanduri V. Senf A. Spalding J. D. Ur-Rehman S. . (2015). The European genome-phenome archive of human data consented for biomedical research. Nat. Genet. 47, 692–695. doi: 10.1038/ng.3312, PMID: 26111507 Larsson S. C. Burgess S. Mason A. M. Michaëlsson K. (2020a). Alcohol consumption and cardiovascular disease: a Mendelian randomization study. Circ. Genom. Precis. Med. 13:e002814. doi: 10.1161/CIRCGEN.119.002814 Larsson S. C. Burgess S. Michaëlsson K. (2019a). Smoking and stroke: a Mendelian randomization study. Ann. Neurol. 86, 468–471. doi: 10.1002/ana.25534 Larsson S. C. Gill D. Mason A. M. Jiang T. Bäck M. Butterworth A. S. . (2020b). Lipoprotein(a) in Alzheimer, atherosclerotic, cerebrovascular, thrombotic, and valvular disease: Mendelian randomization investigation. Circulation 141, 1826–1828. doi: 10.1161/CIRCULATIONAHA.120.045826 Larsson S. C. Markus H. S. (2019). Genetic liability to insomnia and cardiovascular disease risk. Circulation 140, 796–798. doi: 10.1161/CIRCULATIONAHA.119.041830, PMID: 31422675 Larsson S. C. Mason A. M. Bäck M. Klarin D. Damrauer S. M. Michaëlsson K. . (2020c). Genetic predisposition to smoking in relation to 14 cardiovascular diseases. Eur. Heart J. 41, 3304–3310. doi: 10.1093/eurheartj/ehaa193, PMID: 32300774 Larsson S. C. Scott R. A. Traylor M. Langenberg C. C. Hindy G. Melander O. . (2017). Type 2 diabetes, glucose, insulin, BMI, and ischemic stroke subtypes: Mendelian randomization study. Neurology 89, 454–460. doi: 10.1212/WNL.0000000000004173, PMID: 28667182 Larsson S. C. Traylor M. Burgess S. Boncoraglio G. B. Jern C. Michaëlsson K. . (2019b). Serum magnesium and calcium levels in relation to ischemic stroke: Mendelian randomization study. Neurology 92, E944–E950. doi: 10.1212/WNL.0000000000007001 Larsson S. C. Traylor M. Markus H. S. (2018a). Circulating vitamin K1 levels in relation to ischemic stroke and its subtypes: a Mendelian randomization study. Nutrients 10:1575. doi: 10.3390/nu10111575, PMID: 30366361 Larsson S. C. Traylor M. Mishra A. Howson J. M. M. Michaëlsson K. Markus H. S. (2018b). Serum 25-hydroxyvitamin D concentrations and ischemic stroke and its subtypes a Mendelian randomization study. Stroke 49, 2508–2511. doi: 10.1161/STROKEAHA.118.022242 Lin J. Wang Y. Wang Y. Pan Y. (2020). Inflammatory biomarkers and risk of ischemic stroke and subtypes: a 2-sample Mendelian randomization study. Neurol. Res. 42, 118–125. doi: 10.1080/01616412.2019.1710404, PMID: 31900081 Liu J. Rutten-Jacobs L. Liu M. Markus H. S. Traylor M. (2018). Causal impact of type 2 diabetes mellitus on cerebral small vessel disease: a Mendelian randomization analysis. Stroke 49, 1325–1331. doi: 10.1161/STROKEAHA.117.020536, PMID: 29686024 Lu H. Wu P. F. Li R. Z. Zhang W. Huang G. X. (2020). Sleep duration and stroke: a Mendelian randomization study. Front. Neurol. 11:976. doi: 10.3389/fneur.2020.00976, PMID: 33117250 Ma C. Gurol M. E. Huang Z. Lichtenstein A. H. Wang X. Wang Y. . (2019). Low-density lipoprotein cholesterol and risk of intracerebral hemorrhage. Neurology 93, E445–E457. doi: 10.1212/WNL.0000000000007853 Marini S. Merino J. Montgomery B. E. Malik R. Sudlow C. L. Dichgans M. . (2020). Mendelian randomization study of obesity and cerebrovascular disease. Ann. Neurol. 87, 516–524. doi: 10.1002/ana.25686, PMID: 31975536 Marouli E. Kus A. Del Greco M. F. Chaker L. Peeters R. Teumer A. . (2020). Thyroid function affects the risk of stroke via atrial fibrillation: a Mendelian randomization study. J. Clin. Endocrinol. Metab. 105, 2634–2641. doi: 10.1210/clinem/dgaa239, PMID: 32374820 Mokry L. E. Ahmad O. Forgetta V. Thanassoulis G. Richards J. B. (2015). Mendelian randomisation applied to drug development in cardiovascular disease: a review. J. Med. Genet. 52, 71–79. doi: 10.1136/jmedgenet-2014-102438, PMID: 25515070 Morrison J. Knoblauch N. Marcus J. H. Stephens M. He X. (2020). Mendelian randomization accounting for correlated and uncorrelated pleiotropic effects using genome-wide summary statistics. Nat. Genet. 52, 740–747. doi: 10.1038/s41588-020-0631-4, PMID: 32451458 Nagai A. Hirata M. Kamatani Y. Muto K. Matsuda K. Kiyohara Y. . (2017). Overview of the BioBank Japan project: study design and profile. J. Epidemiol. 27, S2–S8. doi: 10.1016/j.je.2016.12.005, PMID: 28189464 Nelson M. R. Tipney H. Painter J. L. Shen J. Nicoletti P. Shen Y. . (2015). The support of human genetic evidence for approved drug indications. Nat. Genet. 47, 856–860. doi: 10.1038/ng.3314, PMID: 26121088 Pan Y. Li H. Wang Y. Meng X. Wang Y. (2019). Causal effect of Lp(a) [lipoprotein(a)] level on ischemic stroke and alzheimer disease a Mendelian randomization study. Stroke 50, 3532–3539. doi: 10.1161/STROKEAHA.119.026872, PMID: 31597550 Peters S. A. E. Huxley R. R. Woodward M. (2013). Smoking as a risk factor for stroke in women compared with men: a systematic review and meta-analysis of 81 cohorts, including 3 980 359 individuals and 42 401 strokes. Stroke 44, 2821–2828. doi: 10.1161/STROKEAHA.113.002342, PMID: 23970792 Powers W. J. Rabinstein A. A. Ackerson T. Adeoye O. M. Bambakidis N. C. Becker K. . (2019). Guidelines for the early management of patients with acute ischemic stroke: 2019 update to the 2018 guidelines for the early management of acute ischemic stroke a guideline for healthcare professionals from the American heart association/American stroke association. Stroke 50, e344–e418. doi: 10.1161/STR.0000000000000211 Qian Y. Ye D. Huang H. Wu D. J. H. Zhuang Y. Jiang X. . (2020). Coffee consumption and risk of stroke: a Mendelian randomization study. Ann. Neurol. 87, 525–532. doi: 10.1002/ana.25693, PMID: 32034791 Qian Y. Ye D. Wu D. J. H. Feng C. Zeng Z. Ye L. . (2019). Role of cigarette smoking in the development of ischemic stroke and its subtypes: a Mendelian randomization study. Clin. Epidemiol. 11, 725–731. doi: 10.2147/CLEP.S215933 Saliba W. Rennert H. S. Barnett-Griness O. Gronich N. Molad J. Rennert G. . (2018). Association of statin use with spontaneous intracerebral hemorrhage: a cohort study. Neurology 91, e400–e409. doi: 10.1212/WNL.0000000000005907, PMID: 29970405 Schmidt A. F. Finan C. Gordillo-Marañón M. Asselbergs F. W. Freitag D. F. Patel R. S. . (2020). Genetic drug target validation using Mendelian randomisation. Nat. Commun. 11:3255. doi: 10.1038/s41467-020-16969-0, PMID: 32591531 Schooling C. M. Luo S. Au Yeung S. L. Thompson D. J. Karthikeyan S. Bolton T. R. . (2018). Genetic predictors of testosterone and their associations with cardiovascular disease and risk factors: a Mendelian randomization investigation. Int. J. Cardiol. 267, 171–176. doi: 10.1016/j.ijcard.2018.05.051, PMID: 29804699 Song L. Sun J. Söderholm M. Melander O. Orho-Melander M. Nilsson J. . (2020). Association of TIM-1 (T-cell immunoglobulin and mucin domain 1) with incidence of stroke. Arterioscler. Thromb. Vasc. Biol. 40, 1777–1786. doi: 10.1161/ATVBAHA.120.314269, PMID: 32460577 Sposato L. A. Cerasuolo J. O. Cipriano L. E. Fang J. Fridman S. Paquet M. . (2018). Atrial fibrillation detected after stroke is related to a low risk of ischemic stroke recurrence. Neurology 90, e924–e931. doi: 10.1212/WNL.0000000000005126, PMID: 29444969 Sposato L. A. Cipriano L. E. Saposnik G. Vargas E. R. Riccio P. M. Hachinski V. (2015). Diagnosis of atrial fibrillation after stroke and transient ischaemic attack: a systematic review and meta-analysis. Lancet Neurol. 14, 377–387. doi: 10.1016/S1474-4422(15)70027-X, PMID: 25748102 Sun L. Clarke R. Bennett D. Guo Y. Walters R. G. Hill M. . (2019). Causal associations of blood lipids with risk of ischemic stroke and intracerebral hemorrhage in Chinese adults. Nat. Med. 25, 569–574. doi: 10.1038/s41591-019-0366-x, PMID: 30858617 Tian D. Wang P. Tang B. Teng X. Li C. Liu X. . (2020). GWAS atlas: a curated resource of genome-wide variant-trait associations in plants and animals. Nucleic Acids Res. 48, D927–D932. doi: 10.1093/nar/gkz828, PMID: 31566222 Titova O. E. Michaëlsson K. Larsson S. C. (2020). Sleep duration and stroke: prospective cohort study and Mendelian randomization analysis. Stroke 51, 3279–3285. doi: 10.1161/STROKEAHA.120.029902, PMID: 32895015 Valdes-Marquez E. Parish S. Clarke R. Stari T. Worrall B. B. Hopewell J. C. (2019). Relative effects of LDL-C on ischemic stroke and coronary disease: a Mendelian randomization study. Neurology 92, E1176–E1187. doi: 10.1212/WNL.0000000000007091 Verbanck M. Chen C. Y. Neale B. Do R. (2018). Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat. Genet. 50, 693–698. doi: 10.1038/s41588-018-0099-7, PMID: 29686387 Walker V. M. Kehoe P. G. Martin R. M. Davies N. M. (2020). Repurposing antihypertensive drugs for the prevention of Alzheimer’s disease: a Mendelian randomization study. Int. J. Epidemiol. 49, 1132–1140. doi: 10.1093/ije/dyz155, PMID: 31335937 Wang M. Bai Y. Wang Z. Zhang Z. Liu D. Lian X. (2021). Higher tea consumption is associated with decreased risk of small vessel stroke. Clin. Nutr. 40, 1430–1435. doi: 10.1016/j.clnu.2020.08.039, PMID: 32943239 Wang X. Dong Y. Qi X. Huang C. Hou L. (2013). Cholesterol levels and risk of hemorrhagic stroke: a systematic review and meta-analysis. Stroke 44, 1833–1839. doi: 10.1161/STROKEAHA.113.001326, PMID: 23704101 Wang Z. Yang F. Ma M. Bao Q. Shen J. Ye F. . (2020). The impact of growth differentiation factor 15 on the risk of cardiovascular diseases: two-sample Mendelian randomization study. BMC Cardiovasc. Disord. 20:462. doi: 10.1186/s12872-020-01744-2, PMID: 33115406 Wong K. M. Langlais K. Tobias G. S. Fletcher-Hoppe C. Krasnewich D. Leeds H. S. . (2017). The dbGaP data browser: a new tool for browsing dbGaP controlled-access genomic data. Nucleic Acids Res. 45, D819–D826. doi: 10.1093/nar/gkw1139, PMID: 27899644 Wu C. Wu L. Wang J. Lin L. Li Y. Lu Q. . (2021). Systematic identification of risk factors and drug repurposing options for Alzheimer’s disease. Alzheimer's Dement. 7:e12148. doi: 10.1002/trc2.12148, PMID: 33718584 Yeung S. L. A. Luo S. Schooling C. M. (2018). The impact of glycated hemoglobin (HbA1c) on cardiovascular disease risk: a Mendelian randomization study using UK biobank. Diabetes Care 41, 1991–1997. doi: 10.2337/dc18-0289 Yuan S. Bäck M. Bruzelius M. Mason A. M. Burgess S. Larsson S. (2019). Plasma phospholipid fatty acids, FADS1 and risk of 15 cardiovascular diseases: a Mendelian randomisation study. Nutrients 11:3001. doi: 10.3390/nu11123001, PMID: 31817859 Yuan S. Carter P. Bruzelius M. Vithayathil M. Kar S. Mason A. M. . (2020a). Effects of tumour necrosis factor on cardiovascular disease and cancer: a two-sample Mendelian randomization study. EBioMedicine 59:102956. doi: 10.1016/j.ebiom.2020.102956, PMID: 32805626 Yuan S. Lin A. He Q.-Q. Burgess S. Larsson S. C. (2020b). Circulating interleukins in relation to coronary artery disease, atrial fibrillation and ischemic stroke and its subtypes: a two-sample Mendelian randomization study. Int. J. Cardiol. 313, 99–104. doi: 10.1016/j.ijcard.2020.03.053 Yuan T. Si S. Li Y. Li W. Chen X. Liu C. . (2020c). Roles for circulating polyunsaturated fatty acids in ischemic stroke and modifiable factors: a Mendelian randomization study. Nutr. J. 19:70. doi: 10.1186/s12937-020-00582-4 Zhang X. Wang A. Zhang J. Singh M. Liu D. Zuo Y. . (2020). Association of plasma C-reactive protein with ischaemic stroke: a Mendelian randomization study. Eur. J. Neurol. 27, 565–571. doi: 10.1111/ene.14113 Zhuang Z. Gao M. Yang R. Li N. Liu Z. Cao W. . (2020). Association of physical activity, sedentary behaviours and sleep duration with cardiovascular diseases and lipid profiles: a Mendelian randomization analysis. Lipids Health Dis. 19:86. doi: 10.1186/s12944-020-01257-z

Funding. GF was supported by the National Institute on Aging (K76AG59992), the National Institute of Neurological Disorders and Stroke (R03NS112859), the American Heart Association (18IDDG34280056 and 817874), the Yale Pepper Scholar Award (P30AG021342), and the Neurocritical Care Society Research Fellowship. JNA is supported by the American Heart Association’s Bugher Fellowship in Hemorrhagic Stroke Research.