Multivariable Mendelian Randomization and Mediation

Statins, Type 2 Diabetes, and Body Mass Index: A Univariable and Multivariable Mendelian Randomization Study

CONTEXT

Statins and possibly other lipid modifiers increase type 2 diabetes risk and body mass index (BMI). However, to what extent BMI mediates the diabetogenic effects of lipid modifiers remains unclear.

OBJECTIVE

We used Mendelian randomization (MR) to investigate the effects of commonly used lipid modifiers on type 2 diabetes risk and glycemic traits, and any mediation by BMI.

METHODS

Using established genetic variants to mimic commonly used lipid modifiers (ie, statins, PCSK9 inhibitors, and ezetimibe), we assessed their associations with type 2 diabetes risk, glycated hemoglobin (HbA1c), fasting insulin, fasting glucose, and BMI in the largest relevant genome-wide association studies (GWAS) in people of European ancestry, and where possible, in East Asians. We used multivariable MR to examine the role of lipid modifiers independent of BMI.

RESULTS

Genetically mimicked effects of statins and ezetimibe, but not PCSK9 inhibitors were associated with higher risk of type 2 diabetes (odds ratio [OR] 1.74 [95% CI, 1.49 to 2.03]; 1.92 [1.22 to 3.02]; 1.06 [0.87 to 1.29] per SD reduction in low-density lipoprotein (LDL)-cholesterol). Of these lipid modifiers, only genetic mimics of statins were associated with higher BMI (0.33 SD [0.29 to 0.38] per SD reduction in LDL-cholesterol), which explained 54% of the total effect of statins on type 2 diabetes risk.

CONCLUSION

Higher BMI mediated more than half of the diabetogenic effects of statins, which did not extend to other commonly used lipid modifiers. Further investigations are needed to clarify drug-specific mechanisms underlying the effects of lipid modifiers on type 2 diabetes.

Cardiovascular risk factors mediating the protective effect of education on cervical spondylosis risk

The causal association between education and cervical spondylosis may be mediated partly through risk factors of cardiovascular disease. The identification of the protective effect of education and the evaluation of risk factors will help to optimize disease prevention at both clinical and public health levels. In this study, we applied several different Mendelian randomization (MR) methods to identify which cardiovascular factors underlie the clustering of cervical spondylosis with cardiovascular disease, and the degree to which these mediate an effect of education. Univariable MR analyses provided evidence supporting a protective effect of genetically predicted education on cervical spondylosis risk, and MVMR further identified the direct effect of education level. Our results also provided evidence supporting the detrimental effects of BMI and smoking on cervical spondylosis risk, with evidence that the effect of education is mediated through BMI and smoking. The proportions of the effect of education mediated through BMI and smoking were 12% and 3%, respectively. These findings highlight education, obesity, and smoking as common mechanisms underlying the clustering of cervical spondylosis with risk factors of cardiovascular disease, which might represent clinical and public health targets for reducing multi-morbidity and the burden of these common conditions.

Diabetes and intervertebral disc degeneration: A Mendelian randomization study

INTRODUCTION

Intervertebral disc degeneration (IVDD) is an important contributor of low back pain, which represents one of the most disabling symptoms within the adult population. Recently, increasing evidence suggests the potential association between Type 2 diabetes mellitus (T2DM) and IVDD. However, the causal relationship between these two common diseases remains unclear.

METHODS

We conducted a two-sample Mendelian randomization (MR) analysis to assess the causal association between T2DM and IVDD. Sensitivity analysis was performed to test for heterogeneity and horizontal pleiotropy. Multivariable MR was also conducted to adjust for the effect of BMI on IVDD.

RESULTS

A total of 128 independent single-nucleotide polymorphisms (SNPs) that were significantly associated with T2DM were selected as instrumental variables in univariable MR analysis. Our results showed that patients with T2DM had a higher risk of developing IVDD (OR, 1.069; 95% CI, 1.026-1.115; p = 0.002). The relationship remained stable in sensitive analysis including multivariable MR, which implicated the direct causal effect of T2DM on IVDD (OR, 1.080; 95% CI, 1.041-1.121; p < 0.001) after adjusting for BMI.

CONCLUSIONS

MR analysis indicated a causal effect of T2DM on IVDD, and the effect persisted even when we accounted for the impact of BMI.

Integrative Multi-omics Analysis of Childhood Aggressive Behavior

This study introduces and illustrates the potential of an integrated multi-omics approach in investigating the underlying biology of complex traits such as childhood aggressive behavior. In 645 twins (cases = 42%), we trained single- and integrative multi-omics models to identify biomarkers for subclinical aggression and investigated the connections among these biomarkers. Our data comprised transmitted and two non-transmitted polygenic scores (PGSs) for 15 traits, 78,772 CpGs, and 90 metabolites. The single-omics models selected 31 PGSs, 1614 CpGs, and 90 metabolites, and the multi-omics model comprised 44 PGSs, 746 CpGs, and 90 metabolites. The predictive accuracy for these models in the test (N = 277, cases = 42%) and independent clinical data (N = 142, cases = 45%) ranged from 43 to 57%. We observed strong connections between DNA methylation, amino acids, and parental non-transmitted PGSs for ADHD, Autism Spectrum Disorder, intelligence, smoking initiation, and self-reported health. Aggression-related omics traits link to known and novel risk factors, including inflammation, carcinogens, and smoking.

Aspirin Use and Risk of Alzheimer's Disease: A 2-Sample Mendelian Randomization Study

BACKGROUND

Observational studies have shown inconsistent findings of the relationships between aspirin use and the risk of Alzheimer's disease (AD).

OBJECTIVE

Since residual confounding and reverse causality were challenging issues inherent in observational studies, we conducted a 2-sample Mendelian randomization analysis (MR) to investigate whether aspirin use was causally associated with the risk of AD.

METHODS

We conducted 2-sample MR analyses utilizing summary genetic association statistics to estimate the potential causal relationship between aspirin use and AD. Single-nucleotide variants associated with aspirin use in a genome-wide association study (GWAS) of UK Biobank were considered as genetic proxies for aspirin use. The GWAS summary-level data of AD were derived from a meta-analysis of GWAS data from the International Genomics of Alzheimer's Project (IGAP) stage I.

RESULTS

Univariable MR analysis based on these two large GWAS data sources showed that genetically proxied aspirin use was associated with a decreased risk of AD (Odds Ratio (OR): 0.87; 95%CI: 0.77-0.99). In multivariate MR analyses, the causal estimates remained significant after adjusting for chronic pain, inflammation, heart failure (OR = 0.88, 95%CI = 0.78-0.98), or stroke (OR = 0.87, 95%CI = 0.77-0.99), but was attenuated when adjusting for coronary heart disease, blood pressure, and blood lipids.

CONCLUSION

Findings from this MR analysis suggest a genetic protective effect of aspirin use on AD, possibly influenced by coronary heart disease, blood pressure, and lipid levels.

Blood cell traits and risk of glaucoma: A two-sample mendelian randomization study

Importance: Glaucoma is the second leading cause of blindness in the world. The causal direction and magnitude of the association between blood cell traits and glaucoma is uncertain because of the susceptibility of observational studies to confounding and reverse causation. Objective: To explore whether there is a causal relationship of blood cell traits including white blood cell (WBC) count (WBCC) and its subtypes [basophil cell count (BASO), monocyte cell count (MONO), lymphocyte cell count (LYMPH), eosinophil cell count (EOS), neutrophil cell count (NEUT)], red blood cell (RBC) count (RBCC), red blood distribution width (RDW), platelet count (PLT), and plateletcrit (PCT) on glaucoma risk. Methods: A two-sample Mendelian randomization (MR) analysis was conducted. Genome-wide significant single nucleotide polymorphisms (SNPs) from published genome-wide association studies (GWAS) on human blood cell traits were utilized as exposure instruments and the dataset for outcome was from the GWAS summary data of glaucoma. In the univariable MR analysis, we examined the association between genetic evidence of blood cell traits and glaucoma. To further investigate the potential causal mechanisms underlying the observed association, we performed multivariable MR analysis with three models, taking into account the mediator effect of inflammation and oxidative stress. According to Bonferroni-corrected for the 10 exposures in 3 methods, the MR study yielded a statistically significant p-value of 0.0017. Results: Genetically BASO, PCT, LYMPH, and PLT were potentially positively associated with glaucoma in the European ancestry [BASO: Odds ratio (OR) = 1.00122, 95% confidence interval (CI), 1.00003-1.00242, p = 0.045; PCT: OR = 1.00078, 95% CI, 1.00012-1.00143, p = 0.019; LYMPH: OR = 1.00076, 95% CI, 1.00002-1.00151, p = 0.045; PLT: OR = 1.00065, 95% CI, 1.00006-1.00123, p = 0.030], There was insufficient evidence to support a causal association of MONO, NEUT, EOS, WBCC, RBCC and RDW (MONO: OR = 1.00050, p = 0.098; NEUT: OR = 1.00028, p = 0.524; EOS: OR = 1.00020, p = 0.562; WBCC: OR = 1.00008, p = 0.830; RBCC: OR = 0.99996, p = 0.920; RDW: OR = 0.99987, p = 0.734) with glaucoma. The multivariable MR with model 1, 2, and 3 demonstrated that BASO, PCT, LYMPH, and PLT were still potentially genetically associated with the risk of glaucoma. Conclusion: Our study reveals a genetic predisposition to higher LYMPH, BASO, PLT, and PCT are associated with a higher risk of glaucoma, whereas WBCC, MONO, EOS, NEUT, RBCC, and RDW are not associated with the occurrence of glaucoma. This finding also supports previous observational studies associating immune components with glaucoma, thus provide guidance on the predication and prevention for glaucoma.

Low vitamin D concentrations and BMI are causal factors for primary biliary cholangitis: A mendelian randomization study

Backgrounds

Observational studies have identified associations between smoking, alcohol use, body mass index (BMI), and the levels of vitamin D with primary biliary cholangitis (PBC). However, there was a lack of randomization control studies to estimate the causal relationship. This study was to investigate the causal estimates for the effects of those risk factors on PBC.

Methods

The genetic instrument variants were extracted from genome-wide association studies in European ancestry. Two-sample mendelian randomization (MR) and multivariable mendelian randomization were used to determine genetically causal estimates. Primary analyses consisted of random-effects and fix-mode inverse-variance-weighted methods, followed by secondary sensitivity analyses to verify the results.

Results

Our study showed that BMI was a causal factor for PBC (OR 1.35; 95% CI=1.03-1.77; p=0.029). In addition, we found that serum vitamin D levels had a protective effect on PBC after adjusting for BMI (OR 0.51; 95% CI=0.32-0.84; p=0.007). However, we failed to identify evidence supporting that genetic causal effect of smoking and alcohol intake were associated with PBC in European countries.

Conclusion

Our results enriched findings from previous epidemiology studies and provided evidence from MR that serum vitamin D concentrations and BMI were independent causal factors for PBC, suggesting that ensuing vitamin D sufficiency and healthy lifestyles might be a cost-effective measure for early intervention for PBC.

Causal association between systemic lupus erythematosus and the risk of dementia: A Mendelian randomization study

Introduction

It is well-documented that systemic lupus erythematosus (SLE) is associated with dementia. However, the genetic causality of this association remains unclear. Mendelian randomization (MR) was used to investigate the potential causal relationship between SLE and dementia risk in the current study.

Methods

We selected 45 single nucleotide polymorphisms (SNPs) associated with SLE from publicly available genome-wide association studies (GWAS). Summary level statistics were obtained from the dementia GWAS database. MR estimates were performed using the inverse variance weighted (IVW) method, MR-Egger method and weighted median (WM) method. Cochran's Q test, the intercept of MR-Egger, MR-Pleiotropy Residual Sum and Outlier method, leave-one-out analysis and funnel plot were applied for sensitivity analyses.

Results

No significant causal association was found between SLE and any type of dementia, including Alzheimer's disease, vascular dementia, frontotemporal dementia, and dementia with Lewy bodies. These findings were robust across several sensitivity analyses.

Conclusion

Overall, our findings do not support a causal association between SLE and dementia risk.

Quantifying the role of transcript levels in mediating DNA methylation effects on complex traits and diseases

High-dimensional omics datasets provide valuable resources to determine the causal role of molecular traits in mediating the path from genotype to phenotype. Making use of molecular quantitative trait loci (QTL) and genome-wide association study (GWAS) summary statistics, we propose a multivariable Mendelian randomization (MVMR) framework to quantify the proportion of the impact of the DNA methylome (DNAm) on complex traits that is propagated through the assayed transcriptome. Evaluating 50 complex traits, we find that on average at least 28.3% (95% CI: [26.9%-29.8%]) of DNAm-to-trait effects are mediated through (typically multiple) transcripts in the cis-region. Several regulatory mechanisms are hypothesized, including methylation of the promoter probe cg10385390 (chr1:8'022'505) increasing the risk for inflammatory bowel disease by reducing PARK7 expression. The proposed integrative framework can be extended to other omics layers to identify causal molecular chains, providing a powerful tool to map and interpret GWAS signals.

Women's reproductive traits and ischemic stroke: a two-sample Mendelian randomization study

OBJECTIVE

We conducted a Mendelian randomization (MR) study to disentangle causal associations between women's reproductive behaviors and ischemic stroke (IS) and investigate the roles of two modifiable risk factors (body mass index (BMI) and educational attainment (EA)) in these associations.

METHODS

Using summary-level data from large-scale genome-wide association studies, we performed univariable MR to examine whether there is genetic evidence that women's reproductive traits are causally associated with IS and its subtypes. Multivariable MR and MR mediation analysis were used to investigate whether BMI and EA are common mechanisms or mediators for these associations. A set of sensitivity analyses were conducted to test valid MR assumptions.

RESULTS

We observed consistent and statistically significant associations across female and sex-combined analyses for earlier age at first birth (AFB) and age at first sexual intercourse (AFS) with a higher risk of IS and large-artery atherosclerotic stroke (LAS) risk in the primary analysis. The odds ratios of IS per 1 SD increase in genetically predicted early AFB and AFS were 0.93 (95% CI, 0.86-0.99; p = 0.046) and 0.83 (95% CI, 0.70-0.97, p = 0.020), respectively. Further analyses indicated that BMI played a shared role in AFS and IS/LAS while EA played a shared role in AFS/AFB and IS/LAS as well as a mediator in the path from AFS to IS/LAS.

INTERPRETATION

These findings may inform prevention strategies and interventions directed toward relative women's reproductive behaviors and IS. Future studies are warranted to explore other factors related to EA which are responsible for these causalities.

Blood Pressure Mediated the Effects of Urinary Uromodulin Levels on Myocardial Infarction: a Mendelian Randomization Study

Background:

The causal links between urinary uromodulin (uUMOD) and cardiovascular disease (CVD) are still not clarified.

Methods:

We first assessed the relationship between uUMOD and CVD using bidirectional 2-sample Mendelian randomization. Then, multivariable Mendelian randomization and product of the coefficients methods were used to investigate the role of blood pressure in mediating the effect of uUMOD on CVD.

Results:

1-unit higher uUMOD level was associated with a higher risk of myocardial infarction (MI), with an odds ratio of 1.08 ([95% CI, 1.02–1.14]; P =0.009), while MI was not associated with uUMOD levels in reverse. Our study did not support the causal effects of uUMOD on other CVD outcomes, including coronary artery disease, atrial fibrillation, heart failure, and ischemic stroke. In multivariable Mendelian Randomization, the direct effects of uUMOD on MI were attenuated to null after introducing systolic blood pressure or diastolic blood pressure. Mediation analysis showed that the indirect effect of uUMOD on MI mediated by systolic blood pressure or diastolic blood pressure was 1.05 ([95% CI, 1.04–1.06]; mediation proportion=69%) and 1.07 ([95% CI, 1.05–1.08]; mediation proportion=87%), respectively. Similar results were found in sensitivity analysis based on different sets of genetic instruments.

Conclusions:

Our findings provide evidence for the effect of higher uUMOD on increasing blood pressure, which mediates a consequent effect on MI risk in the general population. Further studies are necessary to verify the associations between uUMOD and other CVD outcomes.

Potential Mediators of Causal Associations of Circulating Triglycerides With Blood Pressure: Evidence From Genetic and Observational Data

Background:

Existing evidence indicates that elevated triglycerides may affect blood pressure, but the underlying mechanisms are not fully understood. Herein, we aim to identify the intermediaries of associations of triglyceride with systolic blood pressure and diastolic blood pressure using the Mendelian randomization (MR) framework.

Methods:

Triglyceride-associated single nucleotide polymorphisms were extracted and used to match phenotypes in PhenoScanner. From the broad spectrum of possible triglyceride-associated traits, potential mediators linking triglyceride to blood pressure were screened out by MR and MR-based mediation analysis. Moreover, cross-sectional observational data of 206 341 adults were used to validate the mediators identified at the genetic level.

Results:

Among the nearly 100 raw phenotypes matched by 313 triglyceride-associated single nucleotide polymorphisms, 39 traits were filtered and integrated into subsequent analysis. By further filtering using MR analysis, only pulse rate and lymphocyte count (LC) were identified as independent mediators. MR-based mediation analysis showed that genetically predicted LC could mediate 9.2% of the association of triglyceride with systolic blood pressure; genetically predicted pulse rate and LC could mediate 18.3% and 17.6% of the association of triglyceride with DBP, respectively. Observational data also support the mediating role of pulse rate and LC.

Conclusions:

The current findings highlighted the mediating role of pulse rate and LC on the causal pathway from triglyceride to blood pressure and may contribute to a better understanding of the pathogenic mechanism by which high triglyceride affects other cardiometabolic factors.

A Mendelian randomization-based exploration of red blood cell distribution width and mean corpuscular volume with risk of hemorrhagic strokes

Red blood cell distribution width (RCDW) and mean corpuscular volume (MCV) are associated with different risk factors for hemorrhagic stroke. However, whether RCDW and MCV are causally related to hemorrhagic stroke remains poorly understood. Therefore, we explored the causality between RCDW/MCV and nontraumatic hemorrhagic strokes using Mendelian randomization (MR) methods. We extracted exposure and outcome summary statistics from the UK Biobank and FinnGen. We evaluated the causality of RCDW/MCV on four outcomes (subarachnoid hemorrhage [SAH], intracerebral hemorrhage [ICH], nontraumatic intracranial hemorrhage [nITH], and a combination of SAH, cerebral aneurysm, and aneurysm operations) using univariable MR (UMR) and multivariable MR (MVMR). We further performed colocalization and mediation analyses. UMR and MVMR revealed that higher genetically predicted MCV is protective of ICH (UMR: odds ratio [OR] = 0.89 [0.8-0.99], p = 0.036; MVMR: OR = 0.87 [0.78-0.98], p = 0.021) and nITH (UMR: OR = 0.89 [0.82-0.97], p = 0.005; MVMR: OR = 0.88 [0.8-0.96], p = 0.004). There were no strong causal associations between RCDW/MCV and any other outcome. Colocalization analysis revealed a shared causal variant between MCV and ICH; it was not reported to be associated with ICH. Proportion mediated via diastolic blood pressure was 3.1% (0.1%,14.3%) in ICH and 3.4% (0.2%,15.8%) in nITH. The study constitutes the first MR analysis on whether genetically elevated RCDW and MCV affect the risk of hemorrhagic strokes. UMR, MVMR, and mediation analysis revealed that MCV is a protective factor for ICH and nITH, which may inform new insights into the treatments for hemorrhagic strokes.

Substance use and common contributors to morbidity: A genetics perspective

Excessive substance use and substance use disorders (SUDs) are common, serious and relapsing medical conditions. They frequently co-occur with other diseases that are leading contributors to disability worldwide. While heavy substance use may potentiate the course of some of these illnesses, there is accumulating evidence suggesting common genetic architectures. In this narrative review, we focus on four heritable medical conditions - cardiometabolic disease, chronic pain, depression and COVID-19, which are commonly overlapping with, but not necessarily a direct consequence of, SUDs. We find persuasive evidence of underlying genetic liability that predisposes to both SUDs and chronic pain, depression, and COVID-19. For cardiometabolic disease, there is greater support for a potential causal influence of problematic substance use. Our review encourages de-stigmatization of SUDs and the assessment of substance use in clinical settings. We assert that identifying shared pathways of risk has high translational potential, allowing tailoring of treatments for multiple medical conditions. FUNDING: SSR acknowledges T29KT0526, T32IR5226 and DP1DA054394; RLK acknowledges AA028292; AA acknowledges DA054869 & K02DA032573. The funders had no role in the conceptualization or writing of the paper.

A putative causality of vitamin D in common diseases: A mendelian randomization study

Backgrounds

Vitamin D is considered as a nutrient protecting individuals against an array of diseases based on observational studies. Such a protective effect, however, has not been demonstrated by randomized controlled trials. This study aims to explore a putative causal role of vitamin D in common diseases through a two-sample Mendelian randomization (MR) framework.

Methods

Circulating vitamin D was predicted by 41 genetic variants discovered in European populations. Common diseases were verified through two ways, using information from Japanese patients of Biobank Japan and using information from European patients of FinnGen project. We additionally validated the results by replacing vitamin D-associated instrumental variables (IVs) of European population with that of an independent Japanese population and of an independent Indian population. Inverse-variance weighted method was used as the primary analytical approach while a series of MR methods including MR-Egger regression, weighted median, maximum likelihood, MR-PRESSO and multivariate MR were adopted to guarantee MR model assumptions and to detect horizontal pleiotropy.

Results

Genetically predicted vitamin D was significantly associated with an increased risk of Graves' disease (OR = 1.71, 95%CI: 1.25-2.33, P = 0.001) and cataract (OR = 1.14, 95%CI: 1.03-1.28, P = 0.016); while with a decreased risk of esophageal cancer (OR = 0.66, 95%CI: 0.46-0.93, P = 0.019). This significant causal link between vitamin D and cataract was validated replacing IVs identified in the European population with those from Japanese population. No notable associations of vitamin D with other diseases were observed.

Conclusions

Our findings indicate a potential causal role of vitamin D in common diseases, which needs further validation.

Mediators of the association between educational attainment and type 2 diabetes mellitus: a two-step multivariable Mendelian randomisation study

AIMS/HYPOTHESIS

Type 2 diabetes mellitus is a major health burden disproportionately affecting those with lower educational attainment (EA). We aimed to obtain causal estimates of the association between EA and type 2 diabetes and to quantify mediating effects of known modifiable risk factors.

METHODS

We applied two-step, two-sample multivariable Mendelian randomisation (MR) techniques using SNPs as genetic instruments for exposure and mediators, thereby minimising bias due to confounding and reverse causation. We leveraged summary data on genome-wide association studies for EA, proposed mediators (i.e. BMI, blood pressure, smoking, television watching) and type 2 diabetes. The total effect of EA on type 2 diabetes was decomposed into a direct effect and indirect effects through multiple mediators. Additionally, traditional mediation analysis was performed in a subset of the National Health and Nutrition Examination Survey 2013-2014.

RESULTS

EA was inversely associated with type 2 diabetes (OR 0.53 for each 4.2 years of schooling; 95% CI 0.49, 0.56). Individually, the largest contributors were BMI (51.18% mediation; 95% CI 46.39%, 55.98%) and television watching (50.79% mediation; 95% CI 19.42%, 82.15%). Combined, the mediators explained 83.93% (95% CI 70.51%, 96.78%) of the EA-type 2 diabetes association. Traditional analysis yielded smaller effects but showed consistent direction and priority ranking of mediators.

CONCLUSIONS/INTERPRETATION

These results support a potentially causal protective effect of EA against type 2 diabetes, with considerable mediation by a number of modifiable risk factors. Interventions on these factors thus have the potential of substantially reducing the burden of type 2 diabetes attributable to low EA.

Fibrinogen in Alzheimer's Disease, Parkinson's Disease and Lewy Body Dementia: A Mendelian Randomization Study

Fibrinogen is reportedly associated with neurodegenerative diseases (NDs), but the underlying causality remains controversial. Using Mendelian randomization (MR), this study aimed to assess the causal association between fibrinogen and Alzheimer’s disease (AD), Parkinson’s disease (PD), and Lewy body dementia (LBD). Genetic variants associated with fibrinogen and γ-fibrinogen were selected and used as instrumental variables. The effect estimates of the main analysis were obtained by inverse-variance weighting (IVW), complemented by sensitivity analyses to verify model assumptions, and multivariable MR was conducted to control for potential pleiotropic effect. Two-step MR was performed to assess the causal association through mediators. The main analysis suggested no causal association between genetically predicted plasma fibrinogen and γ-fibrinogen levels and the risk of AD, PD, and LBD. The effect estimates did not change in the follow-up sensitivity analyses and MVMR. However, the two-step MR analysis provides evidence that fibrinogen may contribute to the risk of AD via CRP levels. There was an inverse effect of adult height levels on the risk of AD. Our results support the effects of fibrinogen on the risk of AD through increasing plasma CRP levels. Our study found no evidence to support the effects of genetically determined fibrinogen and γ-fibrinogen levels on the risk of PD and LBD. Additionally, our findings suggested an inverse association between genetically determined adult height levels and the risk of AD. Future studies are needed to elucidate the underlying mechanisms and their clinical applications.

Molecular Alterations Caused by Alcohol Consumption in the UK Biobank: A Mendelian Randomisation Study

Alcohol consumption is associated with the development of cardiovascular diseases, cancer, and liver disease. The biological mechanisms are still largely unclear. Here, we aimed to use an agnostic approach to identify phenotypes mediating the effect of alcohol on various diseases.

METHODS

We performed an agnostic association analysis between alcohol consumption (red and white wine, beer/cider, fortified wine, and spirits) with over 7800 phenotypes from the UK biobank comprising 223,728 participants. We performed Mendelian randomisation analysis to infer causality. We additionally performed a Phenome-wide association analysis and a mediation analysis between alcohol consumption as exposure, phenotypes in a causal relationship with alcohol consumption as mediators, and various diseases as the outcome.

RESULTS

Of 45 phenotypes in association with alcohol consumption, 20 were in a causal relationship with alcohol consumption. Gamma glutamyltransferase (GGT; β = 9.44; 95% CI = 5.94, 12.93; Pfdr = 9.04 × 10-7), mean sphered cell volume (β = 0.189; 95% CI = 0.11, 0.27; Pfdr = 1.00 × 10-4), mean corpuscular volume (β = 0.271; 95% CI = 0.19, 0.35; Pfdr = 7.09 × 10-10) and mean corpuscular haemoglobin (β = 0.278; 95% CI = 0.19, 0.36; Pfdr = 1.60 × 10-6) demonstrated the strongest causal relationships. We also identified GGT and physical inactivity as mediators in the pathway between alcohol consumption, liver cirrhosis and alcohol dependence.

CONCLUSION

Our study provides evidence of causality between alcohol consumption and 20 phenotypes and a mediation effect for physical activity on health consequences of alcohol consumption.

Interpretation of Mendelian randomization using a single measure of an exposure that varies over time

BACKGROUND

Mendelian randomization (MR) is a powerful tool through which the causal effects of modifiable exposures on outcomes can be estimated from observational data. Most exposures vary throughout the life course, but MR is commonly applied to one measurement of an exposure (e.g. weight measured once between ages 40 and 60 years). It has been argued that MR provides biased causal effect estimates when applied to one measure of an exposure that varies over time.

METHODS

We propose an approach that emphasizes the liability that causes the entire exposure trajectory. We demonstrate this approach using simulations and an applied example.

RESULTS

We show that rather than estimating the direct or total causal effect of changing the exposure value at a given time, MR estimates the causal effect of changing the underlying liability for the exposure, scaled to the effect of the liability on the exposure at that time. As such, results from MR conducted at different time points are expected to differ (unless the effect of the liability on exposure is constant over time), as we illustrate by estimating the effect of body mass index measured at different ages on systolic blood pressure.

CONCLUSION

Univariable MR results should not be interpreted as time-point-specific direct or total causal effects, but as the effect of changing the liability for the exposure. Estimates of how the effects of a genetic variant on an exposure vary over time, together with biological knowledge that provides evidence regarding likely effective exposure periods, are required to interpret time-point-specific causal effects.

Estimation of causal effects of a time-varying exposure at multiple time points through multivariable mendelian randomization

Mendelian Randomisation (MR) is a powerful tool in epidemiology that can be used to estimate the causal effect of an exposure on an outcome in the presence of unobserved confounding, by utilising genetic variants as instrumental variables (IVs) for the exposure. The effect estimates obtained from MR studies are often interpreted as the lifetime effect of the exposure in question. However, the causal effects of some exposures are thought to vary throughout an individual's lifetime with periods during which an exposure has a greater effect on a particular outcome. Multivariable MR (MVMR) is an extension of MR that allows for multiple, potentially highly related, exposures to be included in an MR estimation. MVMR estimates the direct effect of each exposure on the outcome conditional on all the other exposures included in the estimation. We explore the use of MVMR to estimate the direct effect of a single exposure at different time points in an individual's lifetime on an outcome. We use simulations to illustrate the interpretation of the results from such analyses and the key assumptions required. We show that causal effects at different time periods can be estimated through MVMR when the association between the genetic variants used as instruments and the exposure measured at those time periods varies. However, this estimation will not necessarily identify exact time periods over which an exposure has the most effect on the outcome. Prior knowledge regarding the biological basis of exposure trajectories can help interpretation. We illustrate the method through estimation of the causal effects of childhood and adult BMI on C-Reactive protein and smoking behaviour.

Effects of glycemic traits on left ventricular structure and function: a mendelian randomization study

BACKGROUND

Adverse ventricular structure and function is a key pathogenic mechanism of heart failure. Observational studies have shown that both insulin resistance (IR) and glycemic level are associated with adverse ventricular structure and function. However, whether IR and glycemic level are causally associated with cardiac structure and function remains unclear.

METHODS

Genetic variants for IR, fasting insulin, HbA1c, and fasting glucose were selected based on published genome-wide association studies, which included 188,577, 108,557, 123,665, and 133,010 individuals of European ancestry, respectively. Outcome datasets for left ventricular (LV) parameters were obtained from UK Biobank Cardiovascular Magnetic Resonance sub-study (n = 16,923). Mendelian randomization (MR) analyses with the inverse-variance weighted (IVW) method were used for the primary analyses, while weighted median, MR-Egger, and MR-PRESSO were used for sensitivity analyses. Multivariable MR analyses were also conducted to examine the independent effects of glycemic traits on LV parameters.

RESULTS

In the primary IVW MR analyses, per 1-standard deviation (SD) higher IR was significantly associated with lower LV end-diastolic volume (β = - 0.31 ml, 95% confidence interval [CI] - 0.48 to - 0.14 ml; P = 4.20 × 10^-4), lower LV end-systolic volume (β = - 0.34 ml, 95% CI - 0.51 to - 0.16 ml; P = 1.43 × 10^-4), and higher LV mass to end-diastolic volume ratio (β = 0.50 g/ml, 95% CI 0.32 to 0.67 g/ml; P = 6.24 × 10^-8) after Bonferroni adjustment. However, no associations of HbA1c and fasting glucose were observed with any LV parameters. Results from sensitivity analyses were consistent with the main findings, but with a slightly attenuated estimate. Multivariable MR analyses provided further evidence for an independent effect of IR on the adverse changes in LV parameters after controlling for HbA1c.

CONCLUSIONS

Our study suggests that genetic liability to IR rather than those of glycemic levels are associated with adverse changes in LV structure and function, which may strengthen our understanding of IR as a risk factor for heart failure by providing evidence of direct impact on cardiac morphology.

Applying Mendelian randomization to appraise causality in relationships between nutrition and cancer

Dietary factors are assumed to play an important role in cancer risk, apparent in consensus recommendations for cancer prevention that promote nutritional changes. However, the evidence in this field has been generated predominantly through observational studies, which may result in biased effect estimates because of confounding, exposure misclassification, and reverse causality. With major geographical differences and rapid changes in cancer incidence over time, it is crucial to establish which of the observational associations reflect causality and to identify novel risk factors as these may be modified to prevent the onset of cancer and reduce its progression. Mendelian randomization (MR) uses the special properties of germline genetic variation to strengthen causal inference regarding potentially modifiable exposures and disease risk. MR can be implemented through instrumental variable (IV) analysis and, when robustly performed, is generally less prone to confounding, reverse causation and measurement error than conventional observational methods and has different sources of bias (discussed in detail below). It is increasingly used to facilitate causal inference in epidemiology and provides an opportunity to explore the effects of nutritional exposures on cancer incidence and progression in a cost-effective and timely manner. Here, we introduce the concept of MR and discuss its current application in understanding the impact of nutritional factors (e.g., any measure of diet and nutritional intake, circulating biomarkers, patterns, preference or behaviour) on cancer aetiology and, thus, opportunities for MR to contribute to the development of nutritional recommendations and policies for cancer prevention. We provide applied examples of MR studies examining the role of nutritional factors in cancer to illustrate how this method can be used to help prioritise or deprioritise the evaluation of specific nutritional factors as intervention targets in randomised controlled trials. We describe possible biases when using MR, and methodological developments aimed at investigating and potentially overcoming these biases when present. Lastly, we consider the use of MR in identifying causally relevant nutritional risk factors for various cancers in different regions across the world, given notable geographical differences in some cancers. We also discuss how MR results could be translated into further research and policy. We conclude that findings from MR studies, which corroborate those from other well-conducted studies with different and orthogonal biases, are poised to substantially improve our understanding of nutritional influences on cancer. For such corroboration, there is a requirement for an interdisciplinary and collaborative approach to investigate risk factors for cancer incidence and progression.

Deciphering how early life adiposity influences breast cancer risk using Mendelian randomization

Studies suggest that adiposity in childhood may reduce the risk of breast cancer in later life. The biological mechanism underlying this effect is unclear but is likely to be independent of body size in adulthood. Using a Mendelian randomization framework, we investigate 18 hypothesised mediators of the protective effect of childhood adiposity on later-life breast cancer, including hormonal, reproductive, physical, and glycaemic traits. Our results indicate that, while most of the hypothesised mediators are affected by childhood adiposity, only IGF-1 (OR: 1.08 [1.03: 1.15]), testosterone (total/free/bioavailable ~ OR: 1.12 [1.05: 1.20]), age at menopause (OR: 1.05 [1.03: 1.07]), and age at menarche (OR: 0.92 [0.86: 0.99], direct effect) influence breast cancer risk. However, multivariable Mendelian randomization analysis shows that the protective effect of childhood body size remains unaffected when accounting for these traits (ORs: 0.59-0.67). This suggests that none of the investigated potential mediators strongly contribute to the protective effect of childhood adiposity on breast cancer risk individually. It is plausible, however, that several related traits could collectively mediate the effect when analysed together, and this work provides a compelling foundation for investigating other mediating pathways in future studies.

Mendelian randomization study of circulating lipids and biliary tract cancer among East Asians

BACKGROUND

Associations of High-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, total cholesterol (CHL), and triglyceride (TRG) concentrations with risk of biliary tract cancer (BtC) were conflicting in observational studies. We aim to investigate the causal link between circulating lipids and BtC using genetic information.

METHODS

Single nucleotide polymorphisms of the four circulating lipids (n = 34,421) and BtC (418 cases and 159,201 controls) were retrieved from two independent GWAS studies performed in East Asian populations. Two-sample univariate and multivariate Mendelian Randomization (MR) analyses were conducted to determine the causal link between circulating lipids and BtC.

RESULTS

No significant horizontal pleiotropy was detected for all circulating lipids according to the MR-PRESSO global test (P = 0.458, 0.368, 0.522, and 0.587 for HDL, LDL, CHL, and TRG, respectively). No significant evidence of heterogeneity and directional pleiotropy was detected by the Cochran's Q test and MR-Egger regression. Univariate MR estimates from inverse variance weighting method suggested that one standard deviation (1-SD) increase of inverse-normal transformed HDL (OR = 1.38, 95% CI 0.98-1.94), LDL (OR = 1.46, 95% CI 0.96-2.23), and CHL (OR = 1.34, 95% CI 0.83-2.16) were not significantly associated with BtC risk. Whereas 1-SD increase of inverse-normal transformed TRG showed a significantly negative association with BtC risk (OR = 0.48, 95% CI 0.31-0.74). In multivariate MR analyses including all the four lipid traits, we found that 1-SD increase of LDL and TRG was significantly associated with elevated (OR = 1.32, 95% CI 1.04-2.01) and decreased (OR = 0.54, 95% CI 0.42-0.68) risk of BtC, respectively.

CONCLUSION

Circulating lipids, particularly LDL and TRG, may have roles in the development of BtC. However, the results of this study should be replicated in MR with larger GWAS sample sizes for BtC.

Causal Inference with Genetic Data: Past, Present, and Future

The set of methods discussed in this collection has emerged from the convergence of two scientific fields-genetics and causal inference. In this introduction, we discuss relevant aspects of each field and show how their convergence arises from the natural experiments that genetics offer. We present introductory concepts useful to readers unfamiliar with genetically informed methods for causal inference. We conclude that existing applications and foreseeable developments should ensure that we rapidly reap the rewards of this relatively new field, not only in terms of our understanding of human disease and development, but also in terms of tangible translational applications.

Mendelian Randomization Study on the Putative Causal Effects of Omega-3 Fatty Acids on Low Back Pain

Previous observational studies have suggested an important role of omega-3 in low back pain. In the present study, we used a two-sample Mendelian randomization (MR) study to identify the putative causal link between omega-3 and low back pain. A broadly used genome-wide association study (GWAS) (n = 8,866 individuals from European ancestry) was used to select plasma omega-3 genetic instrumental variables (IVs). A previously reported GWAS (4,863 cases and 74,589 controls from European ancestry) for low back pain were used to assess the effect of plasma omega-3 levels on low back pain. MR-egger_intercept, MR-PRESSO, MR_egger, and inverse variance weighted (IVW) in Cochran's Q-test were used to determine the pleiotropy and heterogeneity, respectively. MR-egger, weighted median, IVW, and weighted mode were used to perform MR analysis. Finally, the effect of a single nucleotide polymorphism (SNP) was used to test the SNP bias. We did not find a significant pleiotropy or heterogeneity of all six selected plasma omega-3 genetic IVs in low back pain GWAS. Expectedly, we found that as plasma omega-3 levels genetically increased, the risk of low back pain had a decreased trend using MR-egger (Beta = −0.593, p = 0.228; OR = 0.553) and weighted mode (Beta = −0.251, p = 0.281; OR = 0.778). This reduced trend was further proven by weighted median (Beta = −0.436, p = 0.025; OR = 0.646) and IVW (Beta = −0.366, p = 0.049; OR = 0.694). Our analysis suggested a putative causal link between genetically increased plasma omega-3 levels and the reduced risk of low back pain in European ancestries. Thus, the supplementation of omega-3 may be important for the prevention and treatment of low back pain.

Mendelian randomization

Mendelian randomization (MR) is a term that applies to the use of genetic variation to address causal questions about how modifiable exposures influence different outcomes. The principles of MR are based on Mendel's laws of inheritance and instrumental variable estimation methods, which enable the inference of causal effects in the presence of unobserved confounding. In this Primer, we outline the principles of MR, the instrumental variable conditions underlying MR estimation and some of the methods used for estimation. We go on to discuss how the assumptions underlying an MR study can be assessed and give methods of estimation that are robust to certain violations of these assumptions. We give examples of a range of studies in which MR has been applied, the limitations of current methods of analysis and the outlook for MR in the future. The difference between the assumptions required for MR analysis and other forms of non-interventional epidemiological studies means that MR can be used as part of a triangulation across multiple sources of evidence for causal inference.

Mendelian Randomization: Concepts and Scope

Mendelian randomization (MR) is a method of studying the causal effects of modifiable exposures (i.e., potential risk factors) on health, social, and economic outcomes using genetic variants associated with the specific exposures of interest. MR provides a more robust understanding of the influence of these exposures on outcomes because germline genetic variants are randomly inherited from parents to offspring and, as a result, should not be related to potential confounding factors that influence exposure-outcome associations. The genetic variant can therefore be used as a tool to link the proposed risk factor and outcome, and to estimate this effect with less confounding and bias than conventional epidemiological approaches. We describe the scope of MR, highlighting the range of applications being made possible as genetic data sets and resources become larger and more freely available. We outline the MR approach in detail, covering concepts, assumptions, and estimation methods. We cover some common misconceptions, provide strategies for overcoming violation of assumptions, and discuss future prospects for extending the clinical applicability, methodological innovations, robustness, and generalizability of MR findings.

Schizophrenia polygenic risk is associated with child mental health problems through early childhood adversity: evidence for a gene-environment correlation

Previous studies have shown that schizophrenia polygenic risk predicts a multitude of mental health problems in the general population. Yet it is unclear by which mechanisms these associations arise. Here, we explored a possible gene-environment correlation in the association of schizophrenia polygenic risk with mental health problems via childhood adversity. This study was embedded in the population-based Generation R Study, including N = 1901 participants with genotyping for schizophrenia polygenic risk, maternal reporting of childhood adversity, and Child Behaviour Checklist measurement of mental health problems. Independent replication was attempted in the Avon Longitudinal Study of Parents and Children (ALSPAC; N = 3641). Associations were analysed with Poisson regression and statistical mediation analysis. Higher burden of schizophrenia polygenic risk was associated with greater exposure to childhood adversity (P-value threshold < 0.5: Generation R Study, OR = 1.08, 95%CI 1.02-1.15, P = 0.01; ALSPAC, OR = 1.02, 95%CI 1.01-1.03, P < 0.01). Childhood adversities partly explained the relationship of schizophrenia polygenic risk with emotional, attention, and thought problems (proportion explained, range 5-23%). Direct effects of schizophrenia polygenic risk and adversity on mental health outcomes were also observed. In summary, genetic liability to schizophrenia increased the risk for mental health problems in the general paediatric population through childhood adversity. Although this finding could result from a mediated causal relationship between genotype and mental health, we argue that these observations most likely reflect a gene-environment correlation, i.e. adversities are a marker for the genetic risk that parents transmit to children. These and similar recent findings raise important conceptual questions about preventative interventions aimed at reducing childhood adversities.

Review of Mendelian Randomization Studies on Endometrial Cancer

Endometrial cancer (EC) is a common gynecological cancer. In some parts of the world, the incidence and mortality of EC are on the rise. Understanding the risk factors of EC is necessary to prevent the occurrence of this disease. Observational studies have revealed the association between certain modifiable environmental risk factors and EC risk. However, due to unmeasured confounding, measurement errors, and reverse causality, observational studies sometimes have limited ability to judge robust causal inferences. In recent years, Mendelian randomization (MR) analysis has received extensive attention, providing valuable insights for cancer-related research, and is expected to identify potential therapeutic interventions. In MR analysis, genetic variation (alleles are randomly assigned during meiosis and are usually independent of environmental or lifestyle factors) is used instead of modifiable exposure to study the relationship between risk factors and disease. Therefore, MR analysis can make causal inference about exposure and disease risk. This review briefly describes the key principles and assumptions of MR analysis; summarizes published MR studies on EC; focuses on the correlation between different risk factors and EC risks; and discusses the application of MR methods in EC research. The results of MR studies on EC showed that type 2 diabetes, uterine fibroids, higher body mass index, higher plasminogen activator inhibitor-1 (PAI-1), higher fasting insulin, early insulin secretion, longer telomere length, higher testosterone and higher plasma cortisol levels are associated with increased risk of EC. In contrast, later age of menarche, higher circulatory tumor necrosis factor, higher low-density lipoprotein cholesterol, and higher sex hormone-binding globulin levels are associated with reduced risk of EC. In general, despite some limitations, MR analysis still provides an effective way to explore the causal relationship between different risk factors and EC.

Mendelian randomization study supports the causal association between serum cystatin C and risk of diabetic nephropathy

AIMS

Cystatin C, an inhibitor of cysteine protease, has been used as a biomarker for estimating glomerular filtration rate. However, the causal relation between cystatin C and diabetic nephropathy remains uncertain.

METHODS

We assessed the causal effect of cystatin C together with other five serum biomarkers including KIM-1, GDF-15, TBIL, uric acid, and Scr on diabetic nephropathy by Mendelian randomization (MR) analysis. 234 genetic variants were selected as instrumental variables to evaluate the causal effect of cystatin C (N_GWAS=361194) on diabetic nephropathy (Ncase/Ncontrol up to 3283/210463). Multivariable MR (MVMR) was performed to assess the stability of cystatin C's causal relationship. Two-step MR was used to assess the mediation effect of BMI and SBP.

RESULTS

Among the six serum biomarkers, only cystatin C causally associated with diabetic nephropathy (IVW OR: 1.36, 95%CI [1.15, 1.61]). After adjusting for the potential confounders BMI and SBP, cystatin C maintained its causal effect on the DN (OR: 1.17, 95%CI [1.02, 1.33]), which means that the risk of DN increased by 17% with an approximate 1 standard deviation (SD) increment of serum cystatin C level. Two-step MR results indicated that BMI might mediate the causal effect of cystatin C on diabetic nephropathy.

INTERPRETATION

Our findings discovered that cystatin C was a risk factor for diabetic nephropathy independent of BMI and SBP in diabetes mellitus patients. Future research is required to illustrate the underlying mechanism and prove targeting circulating cystatin C could be a potential therapy method.

Varicose Veins and Risk of Venous Thromboembolic Diseases: A Two-Sample-Based Mendelian Randomization Study

Background

Varicose veins are found to be associated with increased risk of venous thromboembolism (VTE) in many observational studies, but whether varicose veins are causally associated with VTE remains unclear. Therefore, we used a series of Mendelian randomization (MR) methods to investigate that association.

Methods

23 independent single-nucleotide polymorphisms (SNPs) for varicose veins were obtained from the Pan UK Biobank analysis. The outcomes datasets for deep vein thrombosis (DVT), pulmonary embolism (PE) and venous thromboembolism (VTE) were obtained from the FinnGen study. Before analysis, body mass index (BMI) and height were included as confounders in our MR model. Basic MR [inverse-variance weighted (IVW), weight-median, penalized weighted-median and MR-Egger methods] and MR-PRESSO were performed against each outcome using the whole SNPs and SNPs after excluding those associated with confounders. If causal associations were suggested for any outcome, a basic MR validation analysis, a multivariable MR analysis with BMI and height, a Causal Analysis Using Summary Effect estimates (CAUSE), and a two-step MR analysis with BMI and height, would follow.

Results

Using 21 qualified SNPs, the IVW method (OR: 1.173, 95% CI: 1.070-1.286, p < 0.001, FDR = 0.002), the weighted median method (OR: 1.255, 95% CI: 1.106-1.423, p < 0.001, FDR = 0.001), the penalized weighted median method (OR: 1.299, 95% CI: 1.128-1.495, p < 0.001, FDR = 0.001) and the MR-PRESSO (OR: 1.165, 95% CI: 1.067-1.273, p = 0.003, FDR = 0.009) suggested potential causal effect of varicose veins on DVT, but no cause effect was found for PE and VTE. Excluding SNPs associated with confounders yielded similar results. The causal association with DVT was validated using a self-reported DVT cohort (IVW, OR: 1.107, 95% CI: 1.041-1.178, p = 0.001). The causal association maintained after adjustment for height (OR = 1.105, 95% CI: 1.028-1.188, p = 0.007), BMI (OR = 1.148, 95% CI: 1.059-1.244, p < 0.001) and them both (OR = 1.104, 95% CI: 1.035-1.177, p = 0.003). The causal association also survived the strict CAUSE (p = 0.018). Finally, in two-step MR, height and BMI were found to have causal effects on both varicose veins and DVT.

Conclusion

Genetically predicted varicose veins may have a causal effect on DVT and may be one of the mediators of obesity and taller height that predispose to DVT.

Fasting Insulin and Risk of Overall and 14 Site-Specific Cancers: Evidence From Genetic Data

Objective

Whether fasting insulin (FI) plays a role in cancer risk remains unclear. This study aimed to investigate the association between FI and cancer risk and to explore its potential mediator role in the association between type 2 diabetes mellitus (T2DM) and cancer.

Methods

Two-sample Mendelian randomization (TSMR) analysis was performed to evaluate the effect of FI on overall and 14 site-specific cancers using genome-wide association study (GWAS) summary-level data from Meta-Analyses of Glucose and Insulin-related traits Consortium (MAGIC) and consortia of 14 site-specific cancers. The primary MR approach was conducted by using the random-effect inverse-variance weighted (IVW) method, and sensitivity analyses were implemented by adopting weighted-median, weighted-mode, MR-Egger, and MR-PRESSO tests. Polygenic risk score analysis was executed by using individual-level data from UK Biobank to validate the findings from TSMR analyses. Multivariable Mendelian randomization (MVMR) was carried out to estimate the mediation effect of FI on the association between T2DM and cancer.

Results

TSMR study suggested that genetically determined high FI levels were associated with increased risk of colorectal cancer (odds ratio (OR) = 1.87, 95% CI: 1.23-2.84, p = 0.003) and endometrial cancer (OR = 1.89, 95% CI: 1.08-3.01, p = 0.008), but not associated with overall cancer risk or the other 12 studied cancer sites. Polygenic risk score analysis successfully replicated the association between genetic liability to high FI levels and the increased risk of colorectal and endometrial cancers. MVMR and MR mediation analyses detected an intermediary effect of FI and quantified that FI mediated 21.3% of the association between T2DM and endometrial cancer.

Conclusions

This study demonstrated that FI levels are associated with the risk of colorectal and endometrial cancers, and FI was found to play an intermediary role in the association between T2DM and endometrial cancer. The associations between FI and other cancers need to be further studied.

Genetic Liability to Sedentary Behavior in Relation to Stroke, Its Subtypes and Neurodegenerative Diseases: A Mendelian Randomization Study

Objective: To investigate the causal association of domain-specific sedentary behaviors with cerebrovascular diseases and neurodegenerative diseases, and the potential mediators among these associations. Methods: Genetic instruments were identified for television watching, computer use and driving behavior from a genome-wide association study including 408,815 subjects. Mendelian randomization (MR) analysis was used to estimate the causal effect of sedentary behaviors on the cerebrovascular diseases and neurodegenerative diseases. Multivariable MR analysis was applied to adjust potential confounding factors, and mediation analysis was conducted to explore potential mediators. Results: Genetically predisposition to 1.5 h/day increase in leisure time watching television was associated with increased risk of all-cause stroke [odds ratio (OR) = 1.32, 95% confidence interval (CI) = 1.15-1.52, p-value for MR-Egger method (P _Egger) = 0.11, I ² = 37%, Cochrane's Q = 212, p-value for Cochran Q test (P _Q) < 0.001], and ischemic stroke (OR = 1.28, 95%CI = 1.10-1.49, P _Egger = 0.04, I ² = 35%, Cochrane's Q = 206, P _Q = 0.002). Interestingly, television watching may decrease the risk of Parkinson's disease (OR = 0.65, 95%CI = 0.50-0.84, P _Egger = 0.47, I ² = 19%, Cochrane's Q = 157, P _Q = 0.04). Television watching was a detrimental factor of cognitive performance (estimate = -0.46, 95%CI = -0.55 - -0.37, P _Egger = 0.001, I ² = 85%, Cochrane's Q = 862, P _Q < 0.001). Sensitivity analyses using leave out method and MR-PRESSO method suggested weak evidence of pleiotropy. Conclusion: We provided genetic evidence for the causal association of television watching with increased risk of all-cause stroke and ischemic stroke, decreased risk of Parkinson's disease, and worse cognitive performance. The results should be interpreted with caution considering the pleiotropy.

Understanding the consequences of educational inequalities on periodontitis: Mendelian randomization study

AIM

Higher educational attainment is associated with a lower risk of periodontitis, but the extent to which this association is causal and mediated by intermediate factors is unclear.

METHODS AND MATERIALS

Using summary data from genetic association studies from up to 1.1 million participants of European descent, univariable and multivariable Mendelian randomization analyses were performed to infer the total effect of educational attainment on periodontitis and to estimate the degree to which income, smoking, alcohol consumption, and body mass index mediate the association.

RESULTS

The odds ratio of periodontitis per one standard deviation increment in genetically predicted education was 0.78 (95% CI: 0.68-0.89). The proportions mediated of the total effect of genetically predicted education on periodontitis were 64%, 35%, 15%, and 46% for income, smoking, alcohol consumption, and body mass index, respectively.

CONCLUSIONS

Using a genetic instrumental variable approach, this study triangulated evidence from existing observational epidemiological studies and suggested that higher educational attainment lowers periodontitis risk. Measures to reduce the burden of educational disparities in periodontitis risk may tackle downstream risk factors, particularly income, smoking, and obesity.

Two-Sample Multivariable Mendelian Randomization Analysis Using R

Mendelian randomization is a framework that uses measured variation in genes for assessing and estimating the causal effect of an exposure on an outcome. Multivariable Mendelian randomization is an extension that can assess the causal effect of multiple exposures on an outcome, and can be advantageous when considering a set (>1) of potentially correlated candidate risk factors in evaluating the causal effect of each on a health outcome, accounting for measured pleiotropy. This can be seen, for example, in determining the causal effects of lipids and cholesterol on type 2 diabetes risk, where the correlated risk factors share genetic predictors. Similar to univariate Mendelian randomization, multivariable Mendelian randomization can be conducted using two-sample summary-level data where the gene-exposure and gene-outcome associations are derived from separate samples from the same underlying population. Here, we present a protocol for conducting a two-sample multivariable Mendelian randomization study using the 'MVMR' package in R and summary-level genetic data. We also provide a protocol for searching and obtaining instruments using available data sources in the 'MRInstruments' R package. Finally, we provide general guidelines and discuss the utility of performing a multivariable Mendelian randomization analysis for simultaneously assessing causality of multiple exposures. © 2021 Wiley Periodicals LLC. Basic Protocol: Performing a two-sample multivariable Mendelian randomization analysis using the 'MVMR' package in R and summarized genetic data Support Protocol 1: Installing the 'MVMR' R package Support Protocol 2: Obtaining instruments from the 'MRInstruments' R package.

Type 2 diabetes mellitus increases risk of erectile dysfunction independent of obesity and dyslipidemia: A Mendelian randomization study

BACKGROUND

The causal effects of individual risk factors of metabolic syndrome on erectile dysfunction (ED) are still unclear.

OBJECTIVES

To evaluate the causal effect of risk factors of metabolic syndrome on ED through Mendelian randomization (MR).

MATERIALS AND METHODS

Data for risk factors were obtained from multiple databases with 173,082-757,601 individuals, and that for ED were collected from a genome-wide association study including 223,805 Europeans. We performed univariate MR analysis using inverse-variance weighted, MR-Egger, weighted-median, weighted mode methods and multivariable MR analysis to evaluate the total and direct causal effects.

RESULTS

The univariable MR supported that type 2 diabetes mellitus (odds ratios [OR] = 1.14, 95% confidence intervals [CI]: 1.08-1.21, p < 0.001) and body mass index (BMI) (OR = 1.27, 95% CI: 1.12-1.44, p < 0.001) were associated with ED. After excluding the SNPs associated with BMI and other risk factors, the results of multivariable MR for T2D (OR = 1.15, 95% CI: 1.05-1.25, p = 0.001) remained consistent. However, the results of multivariable MR provided limited evidence for the causality between BMI and ED (OR = 1.06, 95% CI: 0.88-1.29, p = 0.532). For systolic blood pressure and lipid components (low-density lipoprotein, high-density lipoprotein and triglycerides), both univariable and multivariable MR failed to offer sufficient evidence to confirm their causal effect on ED.

CONCLUSION

T2D showed a direct causal effect on ED independent of obesity and dyslipidemia.

Exploring the Relationship Between Schizophrenia and Cardiovascular Disease: A Genetic Correlation and Multivariable Mendelian Randomization Study

Individuals with schizophrenia have a reduced life-expectancy compared to the general population, largely due to an increased risk of cardiovascular disease (CVD). Clinical and epidemiological studies have been unable to unravel the nature of this relationship. We obtained summary-data of genome-wide-association studies of schizophrenia (N = 130 644), heart failure (N = 977 323), coronary artery disease (N = 332 477), systolic and diastolic blood pressure (N = 757 601), heart rate variability (N = 46 952), QT interval (N = 103 331), early repolarization and dilated cardiomyopathy ECG patterns (N = 63 700). We computed genetic correlations and conducted bi-directional Mendelian randomization (MR) to assess causality. With multivariable MR, we investigated whether causal effects were mediated by smoking, body mass index, physical activity, lipid levels, or type 2 diabetes. Genetic correlations between schizophrenia and CVD were close to zero (-0.02-0.04). There was evidence that liability to schizophrenia causally increases heart failure risk. This effect remained consistent with multivariable MR. There was also evidence that liability to schizophrenia increases early repolarization pattern, largely mediated by BMI and lipids. Finally, there was evidence that liability to schizophrenia increases heart rate variability, a direction of effect contrasting clinical studies. There was weak evidence that higher systolic blood pressure increases schizophrenia risk. Our finding that liability to schizophrenia increases heart failure is consistent with the notion that schizophrenia involves a systemic dysregulation of the body with detrimental effects on the heart. To decrease cardiovascular mortality among individuals with schizophrenia, priority should lie with optimal treatment in early stages of psychosis.

Genome-wide association study of occupational attainment as a proxy for cognitive reserve

Occupational attainment, which represents middle-age cognitive activities, is a known proxy marker of cognitive reserve for Alzheimer's disease. Previous genome-wide association studies (GWAS) have identified numerous genetic variants and revealed the genetic architecture of educational attainment, another marker of cognitive reserve. However, the genetic architecture and heritability for occupational attainment remain elusive. We performed a large-scale GWAS of occupational attainment with 248,847 European individuals from the UK Biobank using the proportional odds logistic mixed model method. In this analysis, we defined occupational attainment using the classified job levels formulated in the UK Standard Occupational Classification system considering the individual professional skill and academic level. We identified 30 significant loci (P < 5 × 10-8); 12 were novel variants, unassociated with other traits. Among them, four lead variants were associated with genes expressed in brain tissues by expression quantitative trait loci mapping from 10 brain regions: rs13002946, rs3741368, rs11654986, and rs1627527. The single nucleotide polymorphism (SNP)-based heritability was estimated to be 8.5% (s.e. = 0.004) and partitioned heritability was enriched in the central nervous system and brain tissues. Genetic correlation analysis showed shared genetic backgrounds between occupational attainment and multiple traits, including education, intelligence, leisure activities, life satisfaction, and neuropsychiatric disorders. In two-sample Mendelian randomization (MR) analysis, we demonstrated that high occupation levels were associated with reduced risk for Alzheimer's disease (OR = 0.78, 95% CI = 0.65-0.92 in inverse variance weighted (IVW) method; OR = 0.73, 95% CI = 0.57-0.92 in the weighted median (WM) method). This causal relationship between occupational attainment and Alzheimer's disease was robust in additional sensitivity analysis that excluded potentially pleiotropic SNPs (OR = 0.72, 95% CI = 0.57-0.91 in the IVW method; OR = 0.72, 95% CI = 0.53-0.97 in the WM method). Multivariable MR confirmed that occupational attainment had an independent effect on the risk for Alzheimer's disease even after taking educational attainment into account (OR = 0.72, 95% CI = 0.54-0.95 in the IVW method; OR = 0.68, 95% CI = 0.48-0.97 in the WM method). Overall, our analyses provide insights into the genetic architecture of occupational attainment and demonstrate that occupational attainment is a potential causal protective factor for Alzheimer's disease as a proxy marker of cognitive reserve.

Blood lipids and risk of colon or rectal cancer: a Mendelian randomization study

PURPOSE

Recent Mendelian randomization (MR) studies derived inconsistent results of blood lipids' effect on colorectal cancer, and whether the blood lipids' effect on colon and rectal cancer is different is still unknown. Here, we sought to answer these questions.

METHODS

Primarily, we employed univariable MR to explore the blood lipids' effect on colon and rectal cancer, including high-density lipoprotein (HDL), low-density lipoprotein (LDL), total cholesterol and triglycerides. Then, multivariable MR was also employed to reassess each blood lipid's effect on the two cancers with adjustment of the other lipids. Reverse MR analysis was adopted to determine whether colon or rectal cancer could affect the levels of blood lipids. The Cochrane's Q value was used to evaluate the heterogeneity, and MR-PRESSO was used to appraise the pleiotropy.

RESULTS

Generally, we did not find any significant result between blood lipids and the colon/rectal cancer after Bonferroni correction in the univariable MR analysis. The multivariable MR analysis also obtained the same results. However, it should be noted that higher total cholesterol level might increase the risk of colon cancer (OR = 1.15 [1.01, 1.31], IVW p value = 0.029) but not rectal cancer (OR = 1.02 [0.85, 1.21], IVW p value = 0.853). Such causal relationship turned insignificant in the multivariable MR. The reverse MR analysis suggested that either colon or rectal cancer could increase the levels of blood lipids.

CONCLUSION

We found no association between blood lipids and risk of colon or rectal cancer, except for a positive association between total cholesterol and colon cancer risk.

The Meaning of "Cause" in Genetics

Causation has multiple distinct meanings in genetics. One reason for this is meaning slippage between two concepts of the gene: Mendelian and molecular. Another reason is that a variety of genetic methods address different kinds of causal relationships. Some genetic studies address causes of traits in individuals, which can only be assessed when single genes follow predictable inheritance patterns that reliably cause a trait. A second sense concerns the causes of trait differences within a population. Whereas some single genes can be said to cause population-level differences, most often these claims concern the effects of many genes. Polygenic traits can be understood using heritability estimates, which estimate the relative influences of genetic and environmental differences to trait differences within a population. Attempts to understand the molecular mechanisms underlying polygenic traits have been developed, although causal inference based on these results remains controversial. Genetic variation has also recently been leveraged as a randomizing factor to identify environmental causes of trait differences. This technique-Mendelian randomization-offers some solutions to traditional epidemiological challenges, although it is limited to the study of environments with known genetic influences.

Triangulating Evidence through the Inclusion of Genetically Informed Designs

Much research effort is invested in attempting to determine causal influences on disease onset and progression to inform prevention and treatment efforts. However, this is often dependent on observational data that are prone to well-known limitations, particularly residual confounding and reverse causality. Several statistical methods have been developed to support stronger causal inference. However, a complementary approach is to use design-based methods for causal inference, which acknowledge sources of bias and attempt to mitigate these through the design of the study rather than solely through statistical adjustment. Genetically informed methods provide a novel and potentially powerful extension to this approach, accounting by design for unobserved genetic and environmental confounding. No single approach will be absent from bias. Instead, we should seek and combine evidence from multiple methodologies that each bring different (and ideally uncorrelated) sources of bias. If the results of these different methodologies align-or triangulate-then we can be more confident in our causal inference. To be truly effective, this should ideally be done prospectively, with the sources of evidence specified in advance, to protect against one final source of bias-our own cognitions, expectations, and fondly held beliefs.

Using Mendelian Randomization to Improve the Design of Randomized Trials

Randomized controlled trials and Mendelian randomization studies are two study designs that provide randomized evidence in human biological and medical research. Both exploit the power of randomization to provide unconfounded estimates of causal effect. However, randomized trials and Mendelian randomization studies have very different study designs and scientific objectives. As a result, despite sometimes being referred to as "nature's randomized trial," a Mendelian randomization study cannot be used to replace a randomized trial but instead provides complementary information. In this review, we explain the similarities and differences between randomized trials and Mendelian randomization studies, and suggest several ways that Mendelian randomization can be used to directly inform and improve the design of randomized trials illustrated with practical examples. We conclude by describing how Mendelian randomization studies can employ the principles of trial design to be framed as "naturally randomized trials" that can provide a template for the design of future randomized trials evaluating therapies directed against genetically validated targets.

Cannabis Use, Pulmonary Function, and Lung Cancer Susceptibility: A Mendelian Randomization Study

INTRODUCTION

Because of widespread use, understanding the pulmonary effects of cannabis use is important; but its role independent from tobacco smoking is yet to be elucidated. We used Mendelian randomization (MR) to assess the effect of genetic liability to lifetime cannabis use and cannabis use disorder on pulmonary function and lung cancer.

METHODS

We used four single nucleotide polymorphisms associated with lifetime cannabis use (p value <5 × 10-8) from a genome-wide association study (GWAS) of 184,765 individuals of European descent from the International Cannabis Consortium, 23andme, and U.K. Biobank as instrumental variables. Seven single nucleotide polymorphisms (p value <5 × 10-8) were selected as instruments for cannabis use disorder from a GWAS meta-analysis of 17,068 European ancestry cases and 357,219 controls of European descent from Psychiatric Genomics Consortium Substance Use Disorders working group, Lundbeck Foundation Initiative for Integrative PsychiatricResearch, and deCode. To assess lung function, GWAS included 79,055 study participants of the SpiroMeta Consortium, and for lung cancer GWAS from the International Lung Cancer Consortium contained 29,266 cases and 56,450 controls.

RESULTS

MR revealed that genetic liability to lifetime cannabis use was associated with increased risk of squamous cell carcinoma (OR = 1.22, 95%, confidence interval = 1.07-1.39, p value = 0.003, q value = 0.025). Pleiotropy-robust methods and positive and negative control analyses did not indicate bias in the primary analysis.

CONCLUSIONS

The findings of this MR analysis suggest evidence for a potential causal association between genetic liability for cannabis use and the risk of squamous cell carcinoma. Triangulating MR and observational studies and addressing orthogonal sources of bias are necessary to confirm this finding.

Causal Inference Methods to Integrate Omics and Complex Traits

Major biotechnological advances have facilitated a tremendous boost to the collection of (gen-/transcript-/prote-/methyl-/metabol-)omics data in very large sample sizes worldwide. Coordinated efforts have yielded a deluge of studies associating diseases with genetic markers (genome-wide association studies) or with molecular phenotypes. Whereas omics-disease associations have led to biologically meaningful and coherent mechanisms, the identified (non-germline) disease biomarkers may simply be correlates or consequences of the explored diseases. To move beyond this realm, Mendelian randomization provides a principled framework to integrate information on omics- and disease-associated genetic variants to pinpoint molecular traits causally driving disease development. In this review, we show the latest advances in this field, flag up key challenges for the future, and propose potential solutions.

Mendelian randomisation for mediation analysis: current methods and challenges for implementation

Mediation analysis seeks to explain the pathway(s) through which an exposure affects an outcome. Traditional, non-instrumental variable methods for mediation analysis experience a number of methodological difficulties, including bias due to confounding between an exposure, mediator and outcome and measurement error. Mendelian randomisation (MR) can be used to improve causal inference for mediation analysis. We describe two approaches that can be used for estimating mediation analysis with MR: multivariable MR (MVMR) and two-step MR. We outline the approaches and provide code to demonstrate how they can be used in mediation analysis. We review issues that can affect analyses, including confounding, measurement error, weak instrument bias, interactions between exposures and mediators and analysis of multiple mediators. Description of the methods is supplemented by simulated and real data examples. Although MR relies on large sample sizes and strong assumptions, such as having strong instruments and no horizontally pleiotropic pathways, our simulations demonstrate that these methods are unaffected by confounders of the exposure or mediator and the outcome and non-differential measurement error of the exposure or mediator. Both MVMR and two-step MR can be implemented in both individual-level MR and summary data MR. MR mediation methods require different assumptions to be made, compared with non-instrumental variable mediation methods. Where these assumptions are more plausible, MR can be used to improve causal inference in mediation analysis.

Age at menarche, age at natural menopause, and risk of rheumatoid arthritis - a Mendelian randomization study

BACKGROUND

Hormonal reproductive factors have been suggested to play an important role in the etiology of rheumatoid arthritis (RA), an autoimmune inflammatory disorder affecting primarily women. We conducted a two-sample Mendelian randomization (MR) study examining three relevant exposures, age at menarche (AAM), age at natural menopause (ANM), and age at first birth (AFB) with the risk of RA.

METHODS

We collected summary statistics from the hitherto largest GWAS conducted in AAM (N = 329,345), ANM (N = 69,360), AFB (N = 251,151), and RA (N_case = 14,361, N_control = 43,923), all of European ancestry. We constructed strong instruments using hundreds of exposure-associated genetic variants and estimated causal relationship through different MR approaches including an inverse-variance weighted method, an MR-Egger regression and a weighted median method. We conducted a multivariable MR to control for pleiotropic effect acting in particular through obesity and socioeconomic status. We also performed important sensitivity analyses to verify model assumptions.

RESULTS

We did not find any evidence in support for a causal association between genetically predicted reproductive factors and risk of RA (OR_{per-SD increment in AAM} = 1.06 [0.98-1.15]; OR_{per-SD increment in ANM} = 1.05 [0.98-1.11], OR _{per-SD increment in AFB} = 0.85 [0.65-1.10]). Results remained consistent after removing palindromic SNPs (OR_{per-SD increment in AAM} = 1.06 [0.97-1.15], OR_{per-SD increment in ANM} = 1.05 [0.98-1.13], OR_{per-SD increment in AFB} = 0.81 [0.61-1.07]) or excluding SNPs associated with potential confounding traits (OR_{per-SD increment in AAM} = 1.03 [0.94-1.12], OR_{per-SD increment in ANM} = 1.04 [0.95-1.14]). No outlying instrument was identified through the leave-one-out analysis.

CONCLUSIONS

Our MR study does not convincingly support a casual effect of reproductive factors, as reflected by age at menarche, age at menopause, and age at first birth, on the development of RA. Despite the largely augmented set of instruments we used, these instruments only explained a modest proportion of phenotypic variance of exposures. Our knowledge regarding this topic is still insufficient and future studies with larger sample size should be designed to replicate or dispute our findings.