Beyond Mendelian randomization: how to interpret evidence of shared genetic predictors

Exploring the Two-Way Link between Migraines and Venous Thromboembolism: A Bidirectional Two-Sample Mendelian Randomization Study

BACKGROUND

 The objective of this study is to utilize Mendelian randomization to scrutinize the mutual causality between migraine and venous thromboembolism (VTE) thereby addressing the heterogeneity and inconsistency that were observed in prior observational studies concerning the potential interrelation of the two conditions.

METHODS

 Employing a bidirectional Mendelian randomization approach, the study explored the link between migraine and VTE, incorporating participants of European descent from a large-scale meta-analysis. An inverse-variance weighted (IVW) regression model, with random-effects, leveraging single nucleotide polymorphisms (SNPs) as instrumental variables was utilized to endorse the mutual causality between migraine and VTE. SNP heterogeneity was evaluated using Cochran's Q-test and to account for multiple testing, correction was implemented using the intercept of the MR-Egger method, and a leave-one-out analysis.

RESULTS

 The IVW model unveiled a statistically considerable causal link between migraine and the development of VTE (odds ratio [OR] = 96.155, 95% confidence interval [CI]: 4.342-2129.458, p = 0.004), implying that migraine poses a strong risk factor for VTE development. Conversely, both IVW and simple model outcomes indicated that VTE poses as a weaker risk factor for migraine (IVW OR = 1.002, 95% CI: 1.000-1.004, p = 0.016). The MR-Egger regression analysis denoted absence of evidence for genetic pleiotropy among the SNPs while the durability of our Mendelian randomization results was vouched by the leave-one-out sensitivity analysis.

CONCLUSION

 The findings of this Mendelian randomization assessment provide substantiation for a reciprocal causative association between migraine and VTE within the European population.

The causal relationship between sarcoidosis and autoimmune diseases: a bidirectional Mendelian randomization study in FinnGen

Background

Sarcoidosis has been considered to be associated with many autoimmune diseases (ADs), but the cause-and-effect relationship between these two diseases has not been fully explored. Therefore, the objective of this study is to explore the possible genetic association between sarcoidosis and ADs.

Methods

We conducted a bidirectional Mendelian randomization (MR) study using genetic variants associated with ADs and sarcoidosis (4,041 cases and 371,255 controls) from the FinnGen study. The ADs dataset comprised 96,150 cases and 281,127 controls, encompassing 44 distinct types of autoimmune-related diseases. Subsequently, we identified seven diseases within the ADs dataset with a case size exceeding 3,500 and performed subgroup analyses on these specific diseases.

Results

The MR evidence supported the causal association of genetic predictors of ADs with an increased risk of sarcoidosis (OR = 1.79, 95% CI = 1.59 to 2.02, P IVW-FE = 1.01 × 10-21), and no reverse causation (OR = 1.05, 95% CI 0.99 to 1.12, P IVW-MRE = 9.88 × 10-2). Furthermore, subgroup analyses indicated that genetic predictors of type 1 diabetes mellitus (T1DM), celiac disease, and inflammatory bowel disease (IBD) were causally linked to an elevated risk of sarcoidosis (All P < 6.25 × 10-3). Conversely, genetic predictors of sarcoidosis showed causal associations with a higher risk of type 1 diabetes mellitus (P < 6.25 × 10-3).

Conclusion

The present study established a positive causal relationship between genetic predictors of ADs (e.g. T1DM, celiac disease, and IBD) and the risk of sarcoidosis, with no evidence of reverse causation.

Genetically predicted body mass index and maternal outcomes of pregnancy: A two-sample Mendelian randomisation study

OBJECTIVE

Observational studies have described associations between obesity and adverse outcomes of pregnancy but observational results are liable to influence by residual confounding. Mendelian randomisation (MR) leverages the 'natural' genetic randomisation to risk of an exposure occurring at allele assortment and conception. Similar to randomisation in a clinical trial, this limits the potential for the influence of confounding.

DESIGN

A two-sample MR study.

SETTING

Summary statistics from published genome wide association studies (GWAS) in European ancestry populations.

POPULATION OR SAMPLE

Instrumental variants for body mass index (BMI) were obtained from a study on 434 794 females.

METHODS

Inverse-variance weighted MR was used to assess the association between BMI and all outcomes. Sensitivity analyses with weighted median and MR-Egger were also performed.

MAIN OUTCOME MEASURES

Female-specific genetic association estimates for outcomes were extracted from the sixth round of analysis of the FINNGEN cohort data.

RESULTS

Higher genetically predicted BMI was associated with higher risk of pre-eclampsia (odds ratio [OR] per standard deviation 1.68, 95% confidence interval [CI] 1.46-1.94, P = 8.74 × 10-13 ), gestational diabetes (OR 1.67, 95% CI 1.46-1.92, P = 5.35 × 10-14 ), polyhydramnios (OR 1.40, 95% CI 1.00-1.96, P = 0.049). There was evidence suggestive of a potential association with higher risk of premature rupture of membranes (OR 1.16, 95% CI 1.00-1.36, P = 0.050) and postpartum depression (OR 1.12, 95% CI 0.99-1.27, P = 0.062).

CONCLUSIONS

Higher genetically predicted BMI is associated with marked increase in risk of pre-eclampsia, gestational diabetes and polyhydramnios. The relation between genetically predicted BMI and premature rupture of membranes and postpartum depression should be assessed in further studies. Our study supports efforts to target BMI as a cardinal risk factor for maternal morbidity in pregnancy.

Ethnic-specific associations between body mass index and gastric cancer: a Mendelian randomization study in European and Korean populations

BACKGROUND

Given the uncertainties surrounding the associations in previous epidemiological studies, we conducted linear and nonlinear Mendelian randomization (MR) studies to evaluate whether body mass index (BMI) associated with gastric cancer (GC) risk in European and Korean.

METHODS

Genome-wide association study-summary statistics were used from the Pan-UK Biobank, the Genetic Investigation of Anthropometric Traits consortium, the K-CHIP consortium, and BioBank Japan. BMI-associated single nucleotide polymorphisms (SNPs) were used as instrumental variables (IVs) in MR to identify the association between BMI and GC. Both linear and nonlinear MR analyses were performed. Sensitivity analyses were also conducted for individuals below or above a BMI of 24 kg/m2.

RESULTS

The study used 22 and 55 SNPs as IVs for BMI in European and Korean populations, respectively. Genetically predicted BMI was positively associated with GC risk in the European population (Odds ratio per 1 kg/m2 increase; 95% CI = 1.17; 1.01-1.36 using simple median method), but no significant association was observed in the Korean population. However, the nonlinear MR identified a U-shaped association between BMI and GC in the Korean population, with both low and high BMIs associated with increased GC risk. A BMI of 24 kg/m2 presented the lowest risk. Sensitivity analyses did not yield any genome-wide significant SNPs.

CONCLUSION

While MR analysis suggests a linear association between BMI and GC in those of European ancestry, nonlinear MR hints at a U-shaped association in Koreans. This suggests the association between BMI and GC risk may vary according to ethnic ancestry.

Does enhanced cognitive performance reduce fracture risk? a Mendelian randomization study

OBJECTIVE

While observational studies have suggested a link between cognitive performance and fracture risk, the causality and site-specific nature are unclear. We applied Mendelian randomization (MR) to elucidate these associations.

METHODS

147 single-nucleotide polymorphisms (SNPs) tied strongly to cognitive performance (p< 5e-8) were selected. We performed MR analysis to investigate the causal relationship between cognitive performance and fractures at specific sites, including the wrist, upper arm, shoulder, ribs, sternum, thoracic spine, lumbar spine, pelvis, femur, leg, and ankle. The primary estimate was determined using the inverse variance-weighted method. Additionally, we examined heterogeneity using the MR Pleiotropy RESidual Sum Outlier test and Cochran Q, and employed MR-Egger regression to identify horizontal pleiotropy.

RESULTS

MR analysis identified a causal association between cognitive performance and fractures at the lumbar-spine-pelvis (odds ratio [OR] = 0.727, 95% CI = 0.552-0.956, p = 0.023), and ribs-sternum-thoracic spine sites (OR = 0.774, 95% CI = 0.615-0.974, p = 0.029). However, no causal association was found for fractures at other sites.

CONCLUSIONS

This study provided evidence of a causal connection between cognitive performance and fracture risk at certain locations. These findings underline the potential of cognitive enhancement strategies as innovative and effective methods for fracture prevention.

The causal relationship between osteoarthritis and bladder cancer: A Mendelian randomization study

OBJECTIVE

The causal association between osteoarthritis (OA) and bladder cancer remains unclear. This Mendelian randomization (MR) study was carried out to assess the potential causal effects of any OA, knee OA and hip OA, and bladder cancer.

METHOD

Genome-wide association study (GWAS) summary data for OA and bladder cancer were obtained in GWAS CATALOG, UK Biobank, and FinnGen Consortium. Inverse-variance weighted (IVW) approach was primarily conducted to evaluate the causal relationships between OA and bladder cancer, and MR-Egger intercept and Cochran's Q test were mainly used to estimate heterogeneity and pleiotropy. MR-PRESSO was used to test the presence of horizontal outliers. Leave-one-out analysis was utilized to ensure the reliability of the results.

RESULTS

A higher genetic predisposition to any OA has a causal association with bladder cancer risk, while neither knee OA nor hip OA is causally linked to bladder cancer. MR-Egger intercept analysis exhibited that any OA and knee OA had no pleiotropic effect on the risk of bladder cancer, and Cochran's Q test showed that any OA, knee OA and hip OA had no heterogeneity on bladder cancer risk. Neither MR PRESSO analysis nor leave-one-out analysis revealed any outlier SNPs.

CONCLUSIONS

This MR study exhibited a positive cause-and-effect relationship between any type of OA and bladder cancer risk, but not between site-specific OA, knee OA and hip OA, and bladder cancer. Attention should be paid to the screening and prevention of bladder cancer in OA patients at any site.

Body Mass Index and the Risk of Rheumatic Disease: Linear and Nonlinear Mendelian Randomization Analyses

OBJECTIVE

Although the association between obesity and risk of rheumatic disease is well established, the precise causal relation has not been conclusively proven. Here, we estimate the causal effect of body mass index (BMI) on the risk of developing 5 different rheumatic diseases.

METHODS

Linear and nonlinear mendelian randomization (MR) were used to estimate the effect of BMI on risk of rheumatic disease, and sex-specific effects were identified. Analyses were performed in 361,952 participants from the UK Biobank cohort for 5 rheumatic diseases: rheumatoid arthritis (n = 8,381 cases), osteoarthritis (n = 87,430), psoriatic arthropathy (n = 933), gout (n = 13,638), and inflammatory spondylitis (n = 4,328).

RESULTS

Using linear MR, we found that 1 SD increase in BMI increases the incidence rate for rheumatoid arthritis (incidence rate ratio [IRR] 1.52 [95% confidence interval (95% CI) 1.36-1.69]), osteoarthritis (IRR 1.49 [95% CI 1.43-1.55]), psoriatic arthropathy (IRR 1.80 [95% CI 1.31-2.48]), gout (IRR 1.73 [95% CI 1.56-1.92]), and inflammatory spondylitis (IRR 1.34 [95% CI 1.14-1.57]) in all individuals. BMI was found to be a stronger risk factor in women compared to men for psoriatic arthropathy (P for sex interaction = 3.3 × 10-4 ) and gout (P for sex interaction = 4.3 × 10-3 ), and the effect on osteoarthritis was stronger in premenopausal compared to postmenopausal women (P = 1.8 × 10-3 ). Nonlinear effects of BMI were identified for osteoarthritis and gout in men, and for gout in women. The nonlinearity for gout was also more extreme in men compared to women (P = 0.03).

CONCLUSION

Higher BMI causes an increased risk for rheumatic disease, an effect that is more pronounced in women for both gout and psoriatic arthropathy. The novel sex- and BMI-specific causal effects identified here provide further insight into rheumatic disease etiology and mark an important step toward personalized medicine.

Assessing the safety of lipid-modifying medications among Chinese adolescents: a drug-target Mendelian randomization study

BACKGROUND

With increasing hypercholesterolemia prevalence in East Asian adolescents, pharmacologic interventions (e.g., HMGCR inhibitors (statins) and PCSK9 inhibitors) may have to be considered although their longer-term safety in the general adolescent population is unclear. This study aims to investigate the longer-term safety of HMGCR inhibitors and PCSK9 inhibitors among East Asian adolescents using genetics.

METHODS

A drug-target Mendelian randomization study leveraging the Global Lipid Genetics Consortium (East Asian, n = 146,492) and individual-level data from Chinese participants in the Biobank clinical follow-up of Hong Kong's "Children of 1997" birth cohort (n = 3443, aged ~ 17.6 years). Safety outcomes (n = 100) included anthropometric and hematological traits, renal, liver, lung function, and other nuclear magnetic resonance metabolomics. Positive control outcomes were cholesterol markers from the "Children of 1997" birth cohort and coronary artery disease from Biobank Japan.

RESULTS

Genetic inhibition of HMGCR and PCSK9 were associated with reduction in cholesterol-related NMR metabolomics, e.g., apolipoprotein B (HMGCR: beta [95% CI], - 1.06 [- 1.52 to - 0.60]; PCSK9: - 0.93 [- 1.56 to - 0.31]) and had the expected effect on the positive control outcomes. After correcting for multiple comparisons (p-value < 0.006), genetic inhibition of HMGCR was associated with lower linoleic acid - 0.79 [- 1.25 to - 0.35]. Genetic inhibition of PCSK9 was not associated with the safety outcomes assessed.

CONCLUSIONS

Statins and PCSK9 inhibitors in East Asian adolescents appeared to be safe based on the outcomes concerned. Larger studies were warranted to verify these findings. This study serves as a proof of principle study to inform the medication safety among adolescents via genetics.

Genetic Evidence Strongly Supports Managing Weight and Blood Pressure in Addition to Glycemic Control in Preventing Vascular Complications in People With Type 2 Diabetes

OBJECTIVE

To investigate the causal association of type 2 diabetes and its components with risk of vascular complications independent of shared risk factors obesity and hypertension and to identify the main driver of this risk.

RESEARCH DESIGN AND METHODS

We conducted Mendelian randomization (MR) using independent genetic variants previously associated with type 2 diabetes, fasting glucose, HbA1c, fasting insulin, BMI, and systolic blood pressure as instrumental variables. We obtained summary-level data for 18 vascular diseases (15 for type 2 diabetes) from FinnGen and publicly available genome-wide association studies as our outcomes. We conducted univariable and multivariable MR, in addition to sensitivity tests to detect and minimize pleiotropic effects.

RESULTS

Univariable MR analysis showed that type 2 diabetes was associated with 9 of 15 outcomes; BMI and systolic blood pressure were associated with 13 and 15 of 18 vascular outcomes, respectively; and fasting insulin was associated with 4 and fasting glucose with 2. No robust association was found for HbA1c instruments. With adjustment for correlated traits in the multivariable test, BMI and systolic blood pressure, consistent causal effects were maintained, while five associations with type 2 diabetes (chronic kidney disease, ischemic heart disease, heart failure, subarachnoid hemorrhage, and intracerebral hemorrhage) were attenuated to null.

CONCLUSIONS

Our findings add strong evidence to support the importance of BMI and systolic blood pressure in the development of vascular complications in people with type 2 diabetes. Such findings strongly support the need for better weight and blood pressure management in type 2 diabetes, independent of glucose lowering, to limit important complications.

Association Between Prekallikrein and Stroke: A Mendelian Randomization Study

Background High plasma prekallikrein was reported to be associated with increased risks of stroke, but the causality for these associations remains unclear. We aimed to investigate the associations of genetically predicted plasma prekallikrein concentrations with all-cause stroke, ischemic stroke, 3 ischemic stroke subtypes, and intracerebral hemorrhage (ICH) using a 2-sample Mendelian randomization approach. Methods and Results Seven independent prekallikrein-related single-nucleotide polymorphisms were identified as genetic instruments for prekallikrein based on a genome-wide association study with 1000 European individuals. The summary statistics for all-cause stroke, ischemic stroke, and ischemic stroke subtypes were obtained from the Multiancestry Genome-wide Association Study of Stroke Consortium with 40 585 cases and 406 111 controls of European ancestry. The summary statistics for ICH were obtained from the ISGC (International Stroke Genetics Consortium) with 1545 ICH cases and 1481 controls of European ancestry. In the main analysis, the inverse-variance weighted method was applied to estimate the associations of plasma prekallikrein concentrations with all-cause stroke, ischemic stroke, ischemic stroke subtypes, and ICH. Genetically predicted high plasma prekallikrein levels were significantly associated with elevated risks of all-cause stroke (odds ratio [OR] per SD increase, 1.04 [95% CI, 1.02-1.06]; P=5.44×10-5), ischemic stroke (OR per SD increase, 1.05 [95% CI, 1.03-1.07]; P=1.42×10-5), cardioembolic stroke (OR per SD increase, 1.08 [95% CI, 1.03-1.12]; P=3.75×10-4), and small vessel stroke (OR per SD increase, 1.11 [95% CI, 1.06-1.17]; P=3.02×10-5). However, no significant associations were observed for genetically predicted prekallikrein concentrations with large artery stroke and ICH. Conclusions This Mendelian randomization study found that genetically predicted high plasma prekallikrein concentrations were associated with increased risks of all-cause stroke, ischemic stroke, cardioembolic stroke, and small vessel stroke, indicating that prekallikrein might have a critical role in the development of stroke.

Causal associations of obstructive sleep apnea with cardiovascular disease: a Mendelian randomization study

STUDY OBJECTIVES

Obstructive sleep apnea (OSA) had been associated with various cardiovascular diseases (CVDs) in observational studies, but causal inferences have not been confirmed. We used the Mendelian randomization (MR) study to explore the potential causal association between OSA with CVDs in the general population.

METHODS

We performed a two-sample MR analysis using five gene-wide significant single-nucleotide polymorphisms associated with OSA at genome-wide significance from the FinnGen study (N = 217 955) and 12 cardiovascular diseases from the UK Biobank and the genetic consortia. The inverse-variance weight was chosen as the primary analysis and was complemented by various sensitivity analyses. The study design applied univariable MR, multivariable MR, and mediation analysis.

RESULTS

MR analyses provide evidence of genetically predicted OSA on the risk of heart failure (odds ratio [OR],1.26; 95% confidence interval [CI],1.08 to 1.47), hypertension (OR,1.24; 95%CI, 1.11 to 1.39) and atrial fibrillation (OR,1.21; 95%CI,1.12 to 1.31). Multivariable MR indicated the adverse effect of OSA on heart failure persisted after adjusting BMI, smoking, drinking, and education (IVW OR,1.13; 95%CI, 1.01 to 1.27). However, the significance of hypertension and atrial fibrillation was dampened. Mediation analyses suggest that the causal association between OSA and heart failure is mediated in part by Apolipoprotein B, with a mediated portion of 9%.

CONCLUSIONS

This study suggested that genetically predicted OSA is a potential causal risk factor for heart failure based on a large-scale population. Nevertheless, further studies regarding ancestral diversity are needed to confirm the causal association between OSA and CVDs.

Observational studies in Alzheimer disease: bridging preclinical studies and clinical trials

Recent high-profile failures of Alzheimer disease treatments at the clinical trial stage have led to renewed efforts to identify and test novel interventions for Alzheimer disease and related dementias (ADRD). In this Perspective, we highlight the importance of including well-designed observational studies as part of these efforts. Observational research is an important cornerstone for gathering evidence on risk factors and causes of ADRD; this evidence can then be combined with data from preclinical studies and randomized controlled trials to inform the development of effective interventions. Observational study designs can be particularly beneficial for hypothesis generation, posing questions that are unethical or impractical for a trial setting, studying life-course associations, research in populations typically not included in trials, and public health surveillance. Here, we discuss each of these situations in the specific context of ADRD research. We also highlight novel approaches to enhance causal inference and provide a timely discussion on how observational epidemiological studies help provide a bridge between preclinical studies and successful interventions for ADRD.

Advancing a causal role of type 2 diabetes and its components in developing macro- and microvascular complications via genetic studies

The role of diabetes in developing microvascular and macrovascular complications has been subject to extensive research. Despite multiple observational and genetic studies, the causal inference of diabetes (and associated risk factors) on those complications remains incomplete. In this review, we focused on type 2 diabetes, as the major form of diabetes, and investigated the evidence of causality provided by observational and genetic studies. We found that genetic studies based on Mendelian randomization provided consistent evidence of causal inference of type 2 diabetes on macrovascular complications; however, the evidence for causal inference on microvascular complications has been somewhat limited. We also noted high BMI could be causal for several diabetes complications, notable given high BMI is commonly upstream of type 2 diabetes and the recent calls to target weight loss more aggressively. We emphasize the need for further studies to identify type 2 diabetes components that mostly drive the risk of those complications. Even so, the genetic evidence summarized broadly concurs with the need for a multifactorial risk reduction approach in type 2 diabetes, including addressing excess adiposity.

Avoiding collider bias in Mendelian randomization when performing stratified analyses

Mendelian randomization (MR) uses genetic variants as instrumental variables to investigate the causal effect of a risk factor on an outcome. A collider is a variable influenced by two or more other variables. Naive calculation of MR estimates in strata of the population defined by a collider, such as a variable affected by the risk factor, can result in collider bias. We propose an approach that allows MR estimation in strata of the population while avoiding collider bias. This approach constructs a new variable, the residual collider, as the residual from regression of the collider on the genetic instrument, and then calculates causal estimates in strata defined by quantiles of the residual collider. Estimates stratified on the residual collider will typically have an equivalent interpretation to estimates stratified on the collider, but they are not subject to collider bias. We apply the approach in several simulation scenarios considering different characteristics of the collider variable and strengths of the instrument. We then apply the proposed approach to investigate the causal effect of smoking on bladder cancer in strata of the population defined by bodyweight. The new approach generated unbiased estimates in all the simulation settings. In the applied example, we observed a trend in the stratum-specific MR estimates at different bodyweight levels that suggested stronger effects of smoking on bladder cancer among individuals with lower bodyweight. The proposed approach can be used to perform MR studying heterogeneity among subgroups of the population while avoiding collider bias.

Understanding the assumptions underlying Mendelian randomization

With the rapidly increasing availability of large genetic data sets in recent years, Mendelian Randomization (MR) has quickly gained popularity as a novel secondary analysis method. Leveraging genetic variants as instrumental variables, MR can be used to estimate the causal effects of one phenotype on another even when experimental research is not feasible, and therefore has the potential to be highly informative. It is dependent on strong assumptions however, often producing biased results if these are not met. It is therefore imperative that these assumptions are well-understood by researchers aiming to use MR, in order to evaluate their validity in the context of their analyses and data. The aim of this perspective is therefore to further elucidate these assumptions and the role they play in MR, as well as how different kinds of data can be used to further support them.

Strengthening Causal Inference in Exposomics Research: Application of Genetic Data and Methods

Advances in technologies to measure a broad set of exposures have led to a range of exposome research efforts. Yet, these efforts have insufficiently integrated methods that incorporate genetic data to strengthen causal inference, despite evidence that many exposome-associated phenotypes are heritable. Objective: We demonstrate how integration of methods and study designs that incorporate genetic data can strengthen causal inference in exposomics research by helping address six challenges: reverse causation and unmeasured confounding, comprehensive examination of phenotypic effects, low efficiency, replication, multilevel data integration, and characterization of tissue-specific effects. Examples are drawn from studies of biomarkers and health behaviors, exposure domains where the causal inference methods we describe are most often applied. Discussion: Technological, computational, and statistical advances in genotyping, imputation, and analysis, combined with broad data sharing and cross-study collaborations, offer multiple opportunities to strengthen causal inference in exposomics research. Full application of these opportunities will require an expanded understanding of genetic variants that predict exposome phenotypes as well as an appreciation that the utility of genetic variants for causal inference will vary by exposure and may depend on large sample sizes. However, several of these challenges can be addressed through international scientific collaborations that prioritize data sharing. Ultimately, we anticipate that efforts to better integrate methods that incorporate genetic data will extend the reach of exposomics research by helping address the challenges of comprehensively measuring the exposome and its health effects across studies, the life course, and in varied contexts and diverse populations. https://doi.org/10.1289/EHP9098.

Mendelian randomization in pharmacogenomics: The unforeseen potentials

Mendelian randomization (MR) is an epidemiological method that uses genetic variants to proxy an exposure predicting its causal association with an outcome. It occupies a valuable niche between observational studies and randomized trials. MR applications expanded lately, facilitated by the availability of big data, to include disease risk causation prediction, supporting evidence of prior observational data, identifying new drug targets, and drug repurposing. Concurrently, the last decade witnessed the growth of pharmacogenomics (PGx) research as a cornerstone in precision medicine. PGx research, conducted at discovery and implementation levels, resulted in validated PGx biomarkers and tests. Despite many clinically relevant PGx associations that could be translated into clinical applications, worldwide implementation is lagging far behind. The current review examines the intersection zones between MR and PGx research. MR can provide supporting evidence that allows generalizing PGx findings supporting its implementation. Interchangeability, PGx research can fuel MR studies with libraries of genetic variants of validated biological relevance. Furthermore, PGx and MR exhibit a synergistic relationship in drug discovery that can accelerate identifying new targets and repurposing old drugs. Interdisciplinary research applied by PGx researchers, epidemiologists with MR experience, and data scientists' collaborations can unlock unforeseen opportunities in accelerating precision medicine acquisition.

The African Female Breast Cancer Epidemiology Study Protocol

Breast cancer is now the commonest cancer in most sub-Saharan African countries. Few studies of the epidemiology and genomics of breast cancer and its molecular subtypes in these countries have been done. The African Female Breast Cancer Epidemiology (AFBRECANE) study, a part of the Human Heredity and Health in Africa (H3Africa) initiative, is designed to study the genomics and epidemiology of breast cancer and its molecular subtypes in Nigerian women. We link recruitment of breast cancer cases at study sites with population-based cancer registries activities to enable ascertainment of the incidence of breast cancer and its molecular subtypes. We use centralized laboratory processing to characterize the histopathological and molecular diagnosis of breast cancer and its subtypes using multiple technologies. By combining genome-wide association study (GWAS) data from this study with that generated from 12,000 women participating in our prospective cohort study of cervical cancer, we conduct GWAS of breast cancer in an entirely indigenous African population. We test associations between dietary intakes and breast cancer and focus on vitamin D which we measure using dietary intakes, serum vitamin D, and Mendelian randomization. This paper describes the AFBRECANE project, its design, objectives and anticipated contributions to knowledge and understanding of breast cancer.

Association between type 2 diabetes and amyotrophic lateral sclerosis

Type 2 diabetes (T2D) and amyotrophic lateral sclerosis (ALS) are associated consistently. However, it is currently unknown whether this association is causal. We aimed to estimate the unconfounded, causal association between T2D on ALS using a two-sample Mendelian randomization approach both in European and East Asian ancestry. Genetic variants strongly associated with T2D and each T2D markers were used to investigate the effect of T2D on ALS risk in European (involving 20,806 ALS cases and 59,804 controls) and East Asian (involving 1234 ALS cases and 2850 controls) ancestry. We found that the OR of ALS per 1 SD increase in T2D was estimated to be 0.96 [95% confidence interval (CI) 0.92–0.996; p = 0.03] in European populations. Similarly, all 8 SNPs were associated with T2D in East Asian ancestry, the OR of ALS per 1 SD increase in T2D was estimated to be 0.83 [95% CI 0.70–0.992; p = 0.04] in East Asian populations. Examining the intercept estimates from MR-Egger regression also leads to the same conclusion, in that horizontal pleiotropy unlikely influences the results in either population. We found that genetically predicted T2D was associated with significantly lower odds of amyotrophic lateral sclerosis both in European and East Asian populations. It is now critical to identify a clear molecular explanation for this association between T2D and ALS and to focus on its potential therapeutic implications.

Expression profiles of GPR21, GPR39, GPR135, and GPR153 orphan receptors in different cancers

Orphan receptors have unknown endogenous ligands, are expressed in different tissues, and participate in various diseases such as diabetes, hypertension and cancer. We studied the expression profiles of GPR21, GPR39, GPR135 and GPR153 orphan receptors in several tumour tissues. Cervical, breast, skin, prostate, and astrocytoma tissues were analysed for orphan receptor gene expression using Real time PCR analysis. GPR39 is over-expressed in cervical and prostate cancer tissues, and GPR21 and GPR135 receptors are significantly decreased in cervical, breast, skin, prostate, and astrocytoma tissues, when compared with healthy human fibroblasts. In conclusion, GPR21 and GPR135 receptor gene expression is reduced in cancerous tissues. GPR39 may have a role in the development and evolution of cervical and prostate cancer. These data suggest these receptors may be alternative molecules for new diagnostic approaches, and the design of novel therapeutics against oncological pathologies.

Estimating causal effects of genetically predicted type 2 diabetes on COVID-19 in the East Asian population

BACKGROUND

Observational studies suggested that type 2 diabetes mellitus (T2DM) was associated with an increased risk of coronavirus disease 2019 (COVID-19). However, Mendelian randomization (MR) studies in the European population failed to find causal associations, partly because T2DM was pleiotropically associated with body mass index (BMI). We aimed to estimate the causal effects of T2DM on COVID-19 outcomes in the East Asian (EAS) population using a two-sample MR approach.

METHODS

We obtained summary statistics from a genome-wide association study (GWAS) that included 433,540 EAS participants as the exposure dataset for T2DM risk and from COVID-19 Host Genetics Initiative GWAS meta-analyses (round 7) of EAS ancestry as the outcome dataset for COVID-19 susceptibility (4,459 cases and 36,121 controls), hospitalization (2,882 cases and 31,200 controls), and severity (794 cases and 4,862 controls). As the main MR analysis, we performed the inverse variance weighted (IVW) method. Moreover, we conducted a series of sensitivity analyses, including IVW multivariable MR using summary statistics for BMI from a GWAS with 158,284 Japanese individuals as a covariate.

RESULTS

The IVW method showed that the risk of T2DM significantly increased the risk of COVID-19 susceptibility (odds ratio [OR] per log (OR) increase in T2DM, 1.11; 95% confidence interval [CI], 1.02-1.20; P = 0.014) and hospitalization (OR, 1.15; 95% CI, 1.04-1.26; P = 0.005), although the risk of severity was only suggestive. Moreover, IVW multivariable MR analysis indicated that the causal effects of T2DM on COVID-19 outcomes were independent of the effect of BMI.

CONCLUSIONS

Our MR study indicated for the first time that genetically predicted T2DM is a risk factor for SARS-CoV-2 infection and hospitalized COVID-19 independent of obesity in the EAS population.

Welch-weighted Egger regression reduces false positives due to correlated pleiotropy in Mendelian randomization

Modern population-scale biobanks contain simultaneous measurements of many phenotypes, providing unprecedented opportunity to study the relationship between biomarkers and disease. However, inferring causal effects from observational data is notoriously challenging. Mendelian randomization (MR) has recently received increased attention as a class of methods for estimating causal effects using genetic associations. However, standard methods result in pervasive false positives when two traits share a heritable, unobserved common cause. This is the problem of correlated pleiotropy. Here, we introduce a flexible framework for simulating traits with a common genetic confounder that generalizes recently proposed models, as well as a simple approach we call Welch-weighted Egger regression (WWER) for estimating causal effects. We show in comprehensive simulations that our method substantially reduces false positives due to correlated pleiotropy while being fast enough to apply to hundreds of phenotypes. We apply our method first to a subset of the UK Biobank consisting of blood traits and inflammatory disease, and then to a broader set of 411 heritable phenotypes. We detect many effects with strong literature support, as well as numerous behavioral effects that appear to stem from physician advice given to people at high risk for disease. We conclude that WWER is a powerful tool for exploratory data analysis in ever-growing databases of genotypes and phenotypes.

Genetically Determined TSH Level Within Reference Range Is Inversely Associated With Alzheimer Disease

CONTEXT

Contradictory findings were reported in observational studies on the association of thyroid function (thyrotropin [TSH] and free thyroxine [FT4] levels) with Alzheimer disease (AD).

OBJECTIVE

This work aims to determine whether genetically determined TSH/FT4 levels within reference range are causally associated with AD.

METHODS

A bidirectional, 2-sample mendelian randomization (MR) study was conducted. With summary statistics from the largest genome-wide association studies (GWAS)/GWAS meta-analysis of TSH level(n ≥ 54 288), FT4 level(n = 49 269), and AD (71 880 cases; 383 378 controls), we used an MR approach to evaluate the bidirectional causal relationship between TSH/FT4 levels and AD. The inverse-variance weighted method was adopted as the main analysis.

RESULTS

One SD increase in genetically determined TSH level within reference range was causally associated with a reduced risk of AD (odds ratio: 0.988; 95% CI, 0.977-0.998). A similar inverse association was observed in sex-specific analysis. The causal association was attenuated after adjustment for atrial fibrillation and blood pressure, suggesting they may mediate the causal pathway. A positive causal effect of AD on TSH level was detected only in male participants. This male-specific feedback loop may explain why the largest cohort study to date (Rotterdam Study) demonstrated a null observational association in men. Null association was observed between FT4 level and AD in both directions.

CONCLUSION

Genetic predisposition to increased TSH level, even within reference range, may lower the risk of AD, with atrial fibrillation, systolic, and diastolic blood pressure as possible mediators. Given the higher magnitude of risk reduction observed in the Rotterdam Study, whether the causal estimates derived from this MR study are underestimated warrants further investigation.

Serum Phosphate, BMI, and Body Composition of Middle-Aged and Older Adults: A Cross-Sectional Association Analysis and Bidirectional Mendelian Randomization Study

BACKGROUND

Observational studies have reported associations between serum phosphate and BMI in specific clinical settings, but the nature of this relation in the general population is unclear.

OBJECTIVES

The aim of this study was twofold: to investigate the association between serum phosphate and BMI and body composition, as well as to explore evidence of causality through a bidirectional one-sample Mendelian randomization (MR) in the population-based Rotterdam Study (RS).

METHODS

Observational associations between phosphate (mg/dL) and BMI, lean mass, and fat percentage (fat%), estimated by DXA, were analyzed using multivariable regression models in 9202 participants aged 45-100 y from 3 RS cohorts. The role of serum leptin was examined in a subgroup of 1089 participants. For MR analyses, allele scores with 6 single-nucleotide polymorphisms (SNPs) for phosphate and 905 SNPs for BMI were constructed in 7983 participants.

RESULTS

Phosphate was inversely associated with BMI in the total population (β: -0.89; 95% CI: -1.17, -0.62), and stronger in women (β: -1.92; 95% CI: -2.20, -1.65) than in men (β: -0.37; 95% CI: -0.68, -0.06) (P-interaction < 0.05). Adjustment for leptin did not change results in men. In women, adjustment for leptin attenuated the association, but it was not abolished (β: -0.94; 95% CI: -1.45, -0.42). Phosphate was inversely associated with fat%, but not with lean mass, in both sexes. MR analyses suggested a causal effect of BMI on serum phosphate (β: -0.01; 95% CI: -0.02, 0.00) but not vice versa.

CONCLUSIONS

Serum phosphate was inversely associated with BMI and fat% in a population-based study of middle-aged and older adults, with a stronger effect in women than in men. Adjusting for leptin attenuated this relation in women only. MR results suggest a causal effect of BMI on phosphate but not vice versa. An underlying sex dimorphism in phosphate homeostasis should be further explored.

Extension of Mendelian Randomization to Identify Earliest Manifestations of Alzheimer Disease: Association of Genetic Risk Score for Alzheimer Disease With Lower Body Mass Index by Age 50 Years

Weight loss or lower body mass index (BMI) could be an early symptom of Alzheimer disease (AD), but when this begins to emerge is difficult to estimate with traditional observational data. In an extension of Mendelian randomization, we leveraged variation in genetic risk for late-onset AD risk to estimate the causal effect of AD on BMI and the earliest ages at which AD-related weight loss (or lower BMI as a proxy) occurs. We studied UK Biobank participants enrolled in 2006-2010, who were without dementia, aged 39-73, with European genetic ancestry. BMI was calculated with measured height/weight (weight (kg)/height (m)2). An AD genetic risk score (AD-GRS) was calculated based on 23 genetic variants. Using linear regressions, we tested the association of AD-GRS with BMI, stratified by decade, and calculated the age of divergence in BMI trends between low and high AD-GRS. AD-GRS was not associated with BMI in 39- to 49-year-olds (β = 0.00, 95% confidence interval (CI): -0.03, 0.03). AD-GRS was associated with lower BMI in 50- to 59-year-olds (β = -0.03, 95% CI: -0.06, -0.01) and 60- to 73-year-olds (β = -0.09, 95% CI:-0.12, -0.07). Model-based BMI age curves for high versus low AD-GRS began to diverge after age 47 years. Sensitivity analyses found no evidence for pleiotropy or survival bias. Longitudinal replication is needed; however, our findings suggest that AD genes might begin to reduce BMI decades prior to dementia diagnosis.

Blood copper and risk of cardiometabolic diseases-A Mendelian randomization study

Observational evidence links higher blood levels of copper with higher risk of cardiovascular diseases. However, whether those associations reflect causal links or can be attributed to confounding is still not fully clear. We investigated causal effects of copper on the risk of cardiometabolic endpoints (stroke, coronary artery disease [CAD] and type 2 diabetes) and cardiometabolic risk factors in two-sample Mendelian randomization (MR) studies. The selection of genetic instruments for blood copper levels relied on meta-analysis of genome-wide association studies in three independent studies (European Prospective Investigation into Cancer and Nutrition-Potsdam study, Prospective investigation of the Vasculature in Uppsala Seniors study, Queensland Institute of Medical Research studies). For the selected instruments, outcome associations were drawn from large public genetic consortia on the respective disease endpoints (MEGASTROKE, Cardiogram, DIAGRAM) and cardiometabolic risk factors. MR results indicate an inverse association for genetically higher copper levels with risk of CAD (odds ratio [95% confidence interval] = 0.92 [0.86–0.99], P = 0.022) and systolic blood pressure (beta [standard error (SE)] = −0.238 [0.121]; P = 0.049). Multivariable MR incorporating copper and systolic blood pressure into one model suggested systolic blood pressure as mediating factor between copper and CAD risk. In contrast to previous observational evidence establishing higher blood copper levels as risk-increasing factor for cardiometabolic diseases, this study suggests that higher levels of genetically predicted copper might play a protective role for the development of CAD and systolic blood pressure.

Educational Attainment Decreases the Risk of COVID-19 Severity in the European Population: A Two-Sample Mendelian Randomization Study

Observational studies have reported that the severity of COVID-19 depends not only on physical conditions but also on socioeconomic status, including educational level. Because educational attainment (EA), which measures the number of years of schooling, is moderately heritable, we investigated the causal association of EA on the risk of COVID-19 severity using the Mendelian randomization (MR) approach. A two-sample MR analysis was performed using publicly available summary-level data sets of genome-wide association studies (GWASs). A total of 235 single-nucleotide polymorphisms (SNPs) were extracted as instrumental variables for the exposure of EA from the Social Science Genetic Association Consortium GWAS summary data of 766,345 participants of European ancestry. The effect of each SNP on the outcome of COVID-19 severity risk was obtained from the GWAS summary data of 1,059,456 participants of European ancestry gathered from the COVID-19 Host Genetics Initiative. Using inverse variance weighted method, our MR study shows that EA was significantly associated with a lower risk of COVID-19 severity (odds ratio per one standard deviation increase in years of schooling, 0.540; 95% confidence interval, 0.376–0.777, P = 0.0009). A series of sensitivity analyses showed little evidence of bias. In conclusion, we show for the first time using a two-sample MR approach the associations between higher EA and the lower risk of COVID-19 severity in the European population. However, the genetic or epidemiological mechanisms underlying the association between EA and the risk of COVID-19 severity remain unknown, and further studies are warranted to validate the MR findings and investigate underlying mechanisms.

Determinants of hepatic insulin clearance - Results from a Mendelian Randomization study

AIMS/HYPOTHESIS

Besides insulin resistance, type 2 diabetes associates with decreased hepatic insulin clearance (HIC). We now tested for causal relationship of HIC to liver fat accumulation or features of the metabolic syndrome.

METHODS

HIC was derived from oral glucose tolerance tests with the "Oral C-peptide and Insulin Minimal Models" (n = 3311). Liver fat was quantified by magnetic resonance spectroscopy (n = 1211). Mendelian Randomization was performed using established single nucleotide polymorphisms (SNPs; 115 for liver fat, 155 alanine-aminotransferase, 37 insulin sensitivity, 37 insulin secretion, 72 fasting insulin, 5285 BMI, 163 visceral fat, 270 waist circumference, 442 triglycerides, 620 HDL-Cholesterol, 193 C-reactive protein, 53 lipodystrophy-like phenotypes).

RESULTS

HIC associated inversely with liver fat (p < 0.003) and insulin sensitivity (p < 0.0001). Both liver fat and HIC were independently associated with insulin sensitivity (p < 0.0001). Neither liver fat nor alanine-aminotransferase were causally linked to HIC, as indicated by Mendelian Randomization (N_{liver fat} = 1054, N_HIC = 2254; N_{alanineaminotranferase} = 1985, N_HIC = 2251). BMI-related SNPs were causally associated with HIC (N_BMI = 2772, N_HIC = 2259, p < 0.001) but not waist circumference-SNPs (N_{SNPs-waist circumference} = 2751, N_HIC = 2280). Genetically determined insulin sensitivity was not causally related to HIC (N_{insulin sensitivity} = 2752, N_HIC = 2286). C-reactive protein and HDL were causally associated with HIC, with higher C-reactive protein and lower HDL leading to higher HIC (N_{C-reactive protein} = 2660, N_HIC = 2240; N_HDL = 2694, N_HIC = 2275).

CONCLUSIONS

This Mendelian Randomization analysis does not support a causal link between hepatic steatosis and HIC. Other components of the metabolic syndrome seem to compensate peripheral hyperinsulinemia by increasing hepatic insulin extraction.

Circulating biomarkers of immunity and inflammation, risk of Alzheimer's disease, and hippocampal volume: a Mendelian randomization study

The aim of this study was to explore the association between genetically predicted circulating levels of immunity and inflammation, and the risk of Alzheimer's disease (AD) and hippocampal volume, by conducting a two-sample Mendelian Randomization Study. We identified 12 markers of immune cells and derived ratios (platelet count, eosinophil count, neutrophil count, basophil count, monocyte count, lymphocyte count, platelet-to-lymphocyte ratio, monocyte-to-lymphocyte ratio, CD4 count, CD8 count, CD4-to-CD8 ratio, and CD56) and 5 signaling molecules (IL-6, fibrinogen, CRP, and Lp-PLA2 activity and mass) as primary exposures of interest. Other genetically available immune biomarkers with a weaker a priori link to AD were considered secondary exposures. Associations with AD were evaluated in The International Genomics of Alzheimer's Project (IGAP) GWAS dataset (21,982 cases; 41,944 controls of European ancestry). For hippocampal volume, we extracted data from a GWAS meta-analysis on 33,536 participants of European ancestry. None of the primary or secondary exposures showed statistically significant associations with AD or with hippocampal volume following P-value correction for multiple comparisons using false discovery rate < 5% (Q-value < 0.05). CD4 count showed the strongest suggestive association with AD (odds ratio 1.32, P < 0.01, Q > 0.05). There was evidence for heterogeneity in the MR inverse variance-weighted meta-analyses as measured by Cochran Q, and weighted median and weighted mode for multiple exposures. Further cluster analyses did not reveal clusters of variants that could influence the risk factor in distinct ways. This study suggests that genetically predicted circulating biomarkers of immunity and inflammation are not associated with AD risk or hippocampal volume. Future studies should assess competing risk, explore in more depth the role of adaptive immunity in AD, in particular T cells and the CD4 subtype, and confirm these findings in other ethnicities.

Causal Effects of Serum Levels of n-3 or n-6 Polyunsaturated Fatty Acids on Coronary Artery Disease: Mendelian Randomization Study

We aimed to investigate the causal effects of n-3 and n-6 polyunsaturated fatty acids (PUFAs) on the risk of coronary artery disease (CAD) through Mendelian randomization (MR) analysis. This MR study utilized a genetic instrument developed from previous genome-wide association studies for various serum n-3 and n-6 PUFA levels. First, we calculated the allele scores for genetic predisposition of PUFAs in individuals of European ancestry in the UK Biobank data (N = 337,129). The allele score-based MR was obtained by regressing the allele scores to CAD risks. Second, summary-level MR was performed with the CARDIoGRAMplusC4D data for CAD (N = 184,305). Higher genetically predicted eicosapentaenoic acid and dihomo-gamma-linolenic acid levels were significantly associated with a lower risk of CAD both in the allele-score-based and summary-level MR analyses. Higher allele scores for linoleic acid level were significantly associated with lower CAD risks, and in the summary-level MR, the causal estimates by the pleiotropy-robust MR methods also indicated that higher linoleic acid levels cause a lower risk of CAD. Arachidonic acid showed significant causal estimates for a higher risk of CAD. This study supports the causal effects of certain n-3 and n-6 PUFA types on the risk of CAD.

Genetically Raised Circulating Bilirubin Levels and Risk of Ten Cancers: A Mendelian Randomization Study

Bilirubin, an endogenous antioxidant, may play a protective role in cancer development. We applied two-sample Mendelian randomization to investigate whether genetically raised bilirubin levels are causally associated with the risk of ten cancers (pancreas, kidney, endometrium, ovary, breast, prostate, lung, Hodgkin's lymphoma, melanoma, and neuroblastoma). The number of cases and their matched controls of European descent ranged from 122,977 and 105,974 for breast cancer to 1200 and 6417 for Hodgkin's lymphoma, respectively. A total of 115 single-nucleotide polymorphisms (SNPs) associated (p < 5 × 10-8) with circulating total bilirubin, extracted from a genome-wide association study in the UK Biobank, were used as instrumental variables. One SNP (rs6431625) in the promoter region of the uridine-diphosphoglucuronate glucuronosyltransferase1A1 (UGT1A1) gene explained 16.9% and the remaining 114 SNPs (non-UGT1A1 SNPs) explained 3.1% of phenotypic variance in circulating bilirubin levels. A one-standarddeviation increment in circulating bilirubin (≈ 4.4 µmol/L), predicted by non-UGT1A1 SNPs, was inversely associated with risk of squamous cell lung cancer and Hodgkin's lymphoma (odds ratio (OR) 0.85, 95% confidence interval (CI) 0.73-0.99, P 0.04 and OR 0.64, 95% CI 0.42-0.99, p 0.04, respectively), which was confirmed after removing potential pleiotropic SNPs. In contrast, a positive association was observed with the risk of breast cancer after removing potential pleiotropic SNPs (OR 1.12, 95% CI 1.04-1.20, p 0.002). There was little evidence for robust associations with the other seven cancers investigated. Genetically raised bilirubin levels were inversely associated with risk of squamous cell lung cancer as well as Hodgkin's lymphoma and positively associated with risk of breast cancer. Further studies are required to investigate the utility of bilirubin as a low-cost clinical marker to improve risk prediction for certain cancers.

Lipid and metabolic syndrome traits in coronary artery disease: a Mendelian randomization study

Mendelian randomization (MR) of lipid traits in CAD has provided evidence for causal associations of LDL-C and TGs in CAD, but many lipid trait genetic variants have pleiotropic effects on other cardiovascular risk factors that may bias MR associations. The goal of this study was to evaluate pleiotropic effects of lipid trait genetic variants and to account for these effects in MR of lipid traits in CAD. We performed multivariable MR using inverse variance-weighted and MR-Egger methods in large (n ≥ 300,000) GWAS datasets. We found that 30% of lipid trait genetic variants have effects on metabolic syndrome traits, including BMI, T2D, and systolic blood pressure (SBP). Nonetheless, in multivariable MR analysis, LDL-C, HDL-C, TGs, BMI, T2D, and SBP are independently associated with CAD, and each of these associations is robust to adjustment for directional pleiotropy. MR at loci linked to direct effects on HDL-C and TGs suggests locus- and mechanism-specific causal effects of these factors on CAD.

Is periodontitis a risk factor for ischaemic stroke, coronary artery disease and subclinical atherosclerosis? A Mendelian randomization study

BACKGROUND AND AIMS

Observational studies have reported an association between periodontitis and cardiovascular disease but whether this association is causal is uncertain. We therefore used Mendelian randomization to test whether periodontitis is causally associated with stroke, coronary artery disease, or subclinical atherosclerosis.

METHODS

A two-sample Mendelian randomization analysis was carried out using five single nucleotide polymorphisms previously associated with periodontitis in genome-wide association studies. Summary data were drawn from MEGASTROKE and combined with de novo analyses of UK Biobank for stroke and its major subtypes (up to 44,221 cases, 739,957 controls) and CARDIoGRAMplusC4D and UK Biobank for coronary artery disease (122,733 cases, 424,528 controls). We used existing data on carotid intima-media thickness in UK Biobank as a marker of subclinical atherosclerosis (N = 22,179). Causal estimates were obtained using inverse-variance weighted Mendelian randomization. Sensitivity analyses were performed using weighted median and MR-Egger approaches.

RESULTS

No association was found between periodontitis and any stroke (odds ratio [OR] per doubling in the odds of periodontitis 0.99, 95% confidence interval [CI] 0.97 to 1.02), ischaemic stroke (OR 1.00, 95% CI 0.97 to 1.03) or its major subtypes (p > 0.4), or coronary artery disease (OR 1.01, 95% CI 0.99 to 1.03). Similarly, we found no association for periodontitis and subclinical atherosclerosis (β -0.002, 95% CI -0.004 to 0.001). These results were consistent across a series of sensitivity analyses.

CONCLUSIONS

These findings provide no robust evidence for a causal relationship between periodontitis and stroke or coronary artery disease. This suggests that associations reported in observational studies may represent confounding.

Association of genetic risk for Alzheimer disease and hearing impairment

OBJECTIVE

To test the hypothesis that incipient Alzheimer disease (AD) may adversely affect hearing and that hearing loss may adversely affect cognition, we evaluated whether genetic variants that increase AD risk also increase problem hearing and genetic variants that increase hearing impairment risk do not influence cognition.

METHODS

UK Biobank participants without dementia ≥56 years of age with Caucasian genetic ancestry completed a Digit Triplets Test of speech-in-noise hearing (n = 80,074), self-reported problem hearing and hearing with background noise (n = 244,915), and completed brief cognitive assessments. A genetic risk score for AD (AD-GRS) was calculated as a weighted sum of 23 previously identified AD-related polymorphisms. A genetic risk score for hearing (hearing-GRS) was calculated using 3 previously identified polymorphisms related to hearing impairment. Using age-, sex-, and genetic ancestry-adjusted logistic and linear regression models, we evaluated whether the AD-GRS predicted poor hearing and whether the hearing-GRS predicted worse cognition.

RESULTS

Poor speech-in-noise hearing (>-5.5-dB speech reception threshold; prevalence 14%) was associated with lower cognitive scores (ß = -1.28; 95% confidence interval [CI] -1.54 to -1.03). Higher AD-GRS was significantly associated with poor speech-in-noise hearing (odds ratio [OR] 1.06; 95% CI 1.01-1.11) and self-reported problems hearing with background noise (OR 1.03; 95% CI 1.00-1.05). Hearing-GRS was not significantly associated with cognitive scores (ß = -0.05; 95% CI -0.17 to 0.07).

CONCLUSIONS

Genetic risk for AD also influences speech-in-noise hearing. We failed to find evidence that genetic risk for hearing impairment affects cognition. AD disease processes or a that shared etiology may cause speech-in-noise difficulty before dementia onset.

Sleep, major depressive disorder, and Alzheimer disease: A Mendelian randomization study

OBJECTIVE

To explore the causal relationships between sleep, major depressive disorder (MDD), and Alzheimer disease (AD).

METHODS

We conducted bidirectional 2-sample Mendelian randomization analyses. Genetic associations were obtained from the largest genome-wide association studies currently available in UK Biobank (n = 446,118), Psychiatric Genomics Consortium (n = 18,759), and International Genomics of Alzheimer's Project (n = 63,926). We used the inverse variance-weighted Mendelian randomization method to estimate causal effects and weighted median and Mendelian randomization-Egger for sensitivity analyses to test for pleiotropic effects.

RESULTS

We found that higher risk of AD was significantly associated with being a "morning person" (odds ratio [OR] 1.01, p = 0.001), shorter sleep duration (self-reported: β = -0.006, p = 1.9 × 10^-4; accelerometer based: β = -0.015, p = 6.9 × 10^-5), less likely to report long sleep (β = -0.003, p = 7.3 × 10^-7), earlier timing of the least active 5 hours (β = -0.024, p = 1.7 × 10^-13), and a smaller number of sleep episodes (β = -0.025, p = 5.7 × 10^-14) after adjustment for multiple comparisons. We also found that higher risk of AD was associated with lower risk of insomnia (OR 0.99, p = 7 × 10^-13). However, we did not find evidence that these abnormal sleep patterns were causally related to AD or for a significant causal relationship between MDD and risk of AD.

CONCLUSION

We found that AD may causally influence sleep patterns. However, we did not find evidence supporting a causal role of disturbed sleep patterns for AD or evidence for a causal relationship between MDD and AD.

Genetic Risk of Alzheimer's Disease and Sleep Duration in Non-Demented Elders

Growing evidence has suggested an association between sleep duration and Alzheimer's disease (AD), but it is unclear if sleep duration is a manifestation of the AD disease process. We studied whether genetic liability for AD predicts sleep duration using a genetic risk score (GRS) for AD (AD-GRS), in 406,536 UK Biobank participants with European ancestry and without dementia at enrollment. Higher AD-GRS score was associated with shorter sleep (b = -0.014, 95% confidence interval [CI] = -0.022 to -0.006), especially in those aged 55+. Using AD-GRS as an instrumental variable for AD diagnosis, incipient AD reduced sleep duration by 1.87 hours (95% CI = 0.96, 2.78). Short sleep duration might be an early marker of AD. ANN NEUROL 2021;89:177-181.

Education, intelligence, and amyotrophic lateral sclerosis: A Mendelian randomization study

OBJECTIVE

To systematically investigate causal relationships between educational attainment, cognitive-related phenotypes, and amyotrophic lateral sclerosis (ALS).

METHODS

Summary statistics from genome-wide association studies for educational attainment, math ability, highest math class taken, cognitive performance, intelligence, and ALS were used. A two-sample Mendelian randomization (MR) design was applied to explore the potential causal associations between them.

RESULTS

Genetically predisposition to longer educational attainment and harder math class taken were associated with significantly lower statistical odds ratio of ALS. For per year increase in education completed there was a 21% lower (95% confidence interval [CI] = 27-14%) in risk for ALS. For per 1-SD increase in the highest math class taken obtained there was a 19% lower (95% CI = 9-28%) in risk for ALS. Genetically predisposition to math ability, cognitive performance, and intelligence did not decrease the risk of ALS.

INTERPRETATION

This study provides genetic support for a causal association between higher educational attainment and a lower risk of ALS. The genes related to these phenotypes are involved in almost all aspects of neuron-to-neuron communication. ALS patients are occasionally accompanied by varying degrees of cognitive impairment. People with greater cognitive reserve may be better able to offset damages of degenerative brain changes associated with dementia or other brain diseases, such as Alzheimer's disease, Parkinson's disease, multiple sclerosis, or ALS.

Desaturase Activity and the Risk of Type 2 Diabetes and Coronary Artery Disease: A Mendelian Randomization Study

Estimated Δ5-desaturase (D5D) and Δ6-desaturase (D6D) are key enzymes in metabolism of polyunsaturated fatty acids (PUFA) and have been associated with cardiometabolic risk; however, causality needs to be clarified. We applied two-sample Mendelian randomization (MR) approach using a representative sub-cohort of the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam Study and public data from DIAbetes Genetics Replication And Meta-analysis (DIAGRAM) and Coronary ARtery DIsease Genome wide Replication and Meta-analysis (CARDIoGRAM) genome-wide association studies (GWAS). Furthermore, we addressed confounding by linkage disequilibrium (LD) as all instruments from FADS1 (encoding D5D) are in LD with FADS2 (encoding D6D) variants. Our univariable MRs revealed risk-increasing total effects of both, D6D and D5D on type 2 diabetes (T2DM) risk; and risk-increasing total effect of D6D on risk of coronary artery disease (CAD). The multivariable MR approach could not unambiguously allocate a direct causal effect to either of the individual desaturases. Our results suggest that D6D is causally linked to cardiometabolic risk, which is likely due to downstream production of fatty acids and products resulting from high D6D activity. For D5D, we found indication for causal effects on T2DM and CAD, which could, however, still be confounded by LD.

Testing for causality between systematically identified risk factors and glioma: a Mendelian randomization study

BACKGROUND

Whilst epidemiological studies have provided evidence of associations between certain risk factors and glioma onset, inferring causality has proven challenging. Using Mendelian randomization (MR), we assessed whether associations of 36 reported glioma risk factors showed evidence of a causal relationship.

METHODS

We performed a systematic search of MEDLINE from inception to October 2018 to identify candidate risk factors and conducted a meta-analysis of two glioma genome-wide association studies (5739 cases and 5501 controls) to form our exposure and outcome datasets. MR analyses were performed using genetic variants to proxy for candidate risk factors. We investigated whether risk factors differed by subtype diagnosis (either glioblastoma (n = 3112) or non-glioblastoma (n = 2411)). MR estimates for each risk factor were determined using multiplicative random effects inverse-variance weighting (IVW). Sensitivity analyses investigated potential pleiotropy using MR-Egger regression, the weighted median estimator, and the mode-based estimator. To increase power, trait-specific polygenic risk scores were used to test the association of a genetically predicated increase in each risk factor with glioma onset.

RESULTS

Our systematic search identified 36 risk factors that could be proxied using genetic variants. Using MR, we found evidence that four genetically predicted traits increased risk of glioma, glioblastoma or non-glioblastoma: longer leukocyte telomere length, liability to allergic disease, increased alcohol consumption and liability to childhood extreme obesity (> 3 standard deviations from the mean). Two traits decreased risk of non-glioblastoma cancers: increased low-density lipoprotein cholesterol (LDLc) and triglyceride levels. Our findings were similar across sensitivity analyses that made allowance for pleiotropy (genetic confounding).

CONCLUSIONS

Our comprehensive investigation provides evidence of a causal link between both genetically predicted leukocyte telomere length, allergic disease, alcohol consumption, childhood extreme obesity, and LDLc and triglyceride levels, and glioma. The findings from our study warrant further research to uncover mechanisms that implicate these traits in glioma onset.

Genetically modulated educational attainment and coronary disease risk

AIMS

Genetic disposition and lifestyle factors are understood as independent components underlying the risk of multiple diseases. In this study, we aim to investigate the interplay between genetics, educational attainment-an important denominator of lifestyle-and coronary artery disease (CAD) risk.

METHODS AND RESULTS

Based on the effect sizes of 74 genetic variants associated with educational attainment, we calculated a 'genetic education score' in 13 080 cases and 14 471 controls and observed an inverse correlation between the score and risk of CAD [P = 1.52 × 10-8; odds ratio (OR) 0.79, 95% confidence interval (CI) 0.73-0.85 for the higher compared with the lowest score quintile]. We replicated in 146 514 individuals from UK Biobank (P = 1.85 × 10-6) and also found strong associations between the 'genetic education score' with 'modifiable' risk factors including smoking (P = 5.36 × 10-23), body mass index (BMI) (P = 1.66 × 10-30), and hypertension (P = 3.86 × 10-8). Interestingly, these associations were only modestly attenuated by adjustment for years spent in school. In contrast, a model adjusting for BMI and smoking abolished the association signal between the 'genetic education score' and CAD risk suggesting an intermediary role of these two risk factors. Mendelian randomization analyses performed with summary statistics from large genome-wide meta-analyses and sensitivity analysis using 1271 variants affecting educational attainment (OR 0.68 for the higher compared with the lowest score quintile; 95% CI 0.63-0.74; P = 3.99 × 10-21) further strengthened these findings.

CONCLUSION

Genetic variants known to affect educational attainment may have implications for a health-conscious lifestyle later in life and subsequently affect the risk of CAD.

Do Cerebral Small Vessel Disease and Multiple Sclerosis Share Common Mechanisms of White Matter Injury?

Supplemental Digital Content is available in the text.

Background and Purpose—

The role of inflammation in ischemic white matter disease is increasingly recognized, and further understanding of the pathophysiology might inform future treatment strategies. Multiple sclerosis (MS) is a chronic autoimmune condition in which inflammation plays a central role that also affects the white matter. We hypothesized that white matter injury might share common mechanisms and used statistical genetics techniques to assess whether having genetically elevated white matter hyperintensity (WMH) volume was associated with increased MS risk.

Methods—

We investigated the genetic association in 2 cohorts with magnetic resonance imaging-quantified ischemic white matter lesion volume (WMH in stroke; n=2797 and UK Biobank; n=8353) and 14 802 cases of MS and 26 703 controls from the International Multiple Sclerosis Genetics Consortium. We further performed individual-level polygenic risk score calculations for MS and measures of structural white matter disease in UK Biobank. Finally, we looked for evidence of overlapping risk across the whole genome.

Results—

There was no association of genetic variants influencing MS with WMH volume using summary statistics in the WMH in stroke cohort (relative risk score =1.014; 95% CI, 0.936–1.110) or in the UK Biobank cohort (relative risk score =1.030; 95% CI, 0.932–1.117). Conversely, assessing the contribution of single nucleotide polymorphisms significantly associated with WMH on the risk of MS there was no significant association (relative risk score =0.930; 95% CI, 0.736–1.191). There were no significant associations between polygenic risk scores calculations; these results were robust to the selection of single nucleotide polymorphisms at a range of significance thresholds. Whole genome analysis did not reveal any overlap of risk between the traits.

Conclusions—

Our results do not provide evidence to suggest a shared mechanism of white matter damage in ischemia and MS. We propose that inflammation acts in distinct pathways because of the differing nature of the primary insult.

An examination of multivariable Mendelian randomization in the single-sample and two-sample summary data settings

BACKGROUND

Mendelian randomization (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 utilizing genetic variants that are instrumental variables (IVs) for the exposure. This has been extended to multivariable MR (MVMR) to estimate the effect of two or more exposures on an outcome.

METHODS AND RESULTS

We use simulations and theory to clarify the interpretation of estimated effects in a MVMR analysis under a range of underlying scenarios, where a secondary exposure acts variously as a confounder, a mediator, a pleiotropic pathway and a collider. We then describe how instrument strength and validity can be assessed for an MVMR analysis in the single-sample setting, and develop tests to assess these assumptions in the popular two-sample summary data setting. We illustrate our methods using data from UK Biobank to estimate the effect of education and cognitive ability on body mass index.

CONCLUSION

MVMR analysis consistently estimates the direct causal effect of an exposure, or exposures, of interest and provides a powerful tool for determining causal effects in a wide range of scenarios with either individual- or summary-level data.

"Mendelian Randomization" Approach in Economic Assessment of Health Conditions

The increased prevalence of non-communicable chronic diseases (NCDs) is reflected in the rising economic burden of health conditions. Observational studies conducted in health economics research are detecting associations of NCDs or related risk factors with economic measures like health insurance, economic inequalities, accessibility of jobs, education, annual income, health expenditure, etc. The inferences of such relationships do not prove causation and are limited to associations which are many times influenced by confounding factors and reverse causation. Mendelian randomization (MR) approach is a useful method for exploring causal relations between modifiable risk factors and measures of health economics. The application of MR in economic assessment of health conditions has been started and is producing fruitful results.

The influence of obesity-related factors in the etiology of renal cell carcinoma-A mendelian randomization study

BACKGROUND

Several obesity-related factors have been associated with renal cell carcinoma (RCC), but it is unclear which individual factors directly influence risk. We addressed this question using genetic markers as proxies for putative risk factors and evaluated their relation to RCC risk in a mendelian randomization (MR) framework. This methodology limits bias due to confounding and is not affected by reverse causation.

METHODS AND FINDINGS

Genetic markers associated with obesity measures, blood pressure, lipids, type 2 diabetes, insulin, and glucose were initially identified as instrumental variables, and their association with RCC risk was subsequently evaluated in a genome-wide association study (GWAS) of 10,784 RCC patients and 20,406 control participants in a 2-sample MR framework. The effect on RCC risk was estimated by calculating odds ratios (ORSD) for a standard deviation (SD) increment in each risk factor. The MR analysis indicated that higher body mass index increases the risk of RCC (ORSD: 1.56, 95% confidence interval [CI] 1.44-1.70), with comparable results for waist-to-hip ratio (ORSD: 1.63, 95% CI 1.40-1.90) and body fat percentage (ORSD: 1.66, 95% CI 1.44-1.90). This analysis further indicated that higher fasting insulin (ORSD: 1.82, 95% CI 1.30-2.55) and diastolic blood pressure (DBP; ORSD: 1.28, 95% CI 1.11-1.47), but not systolic blood pressure (ORSD: 0.98, 95% CI 0.84-1.14), increase the risk for RCC. No association with RCC risk was seen for lipids, overall type 2 diabetes, or fasting glucose.

CONCLUSIONS

This study provides novel evidence for an etiological role of insulin in RCC, as well as confirmatory evidence that obesity and DBP influence RCC risk.

Inferring Causal Relationships Between Risk Factors and Outcomes from Genome-Wide Association Study Data

An observational correlation between a suspected risk factor and an outcome does not necessarily imply that interventions on levels of the risk factor will have a causal impact on the outcome (correlation is not causation). If genetic variants associated with the risk factor are also associated with the outcome, then this increases the plausibility that the risk factor is a causal determinant of the outcome. However, if the genetic variants in the analysis do not have a specific biological link to the risk factor, then causal claims can be spurious. We review the Mendelian randomization paradigm for making causal inferences using genetic variants. We consider monogenic analysis, in which genetic variants are taken from a single gene region, and polygenic analysis, which includes variants from multiple regions. We focus on answering two questions: When can Mendelian randomization be used to make reliable causal inferences, and when can it be used to make relevant causal inferences?

Mendelian randomization: a novel approach for the prediction of adverse drug events and drug repurposing opportunities

Identification of unintended drug effects, specifically drug repurposing opportunities and adverse drug events, maximizes the benefit of a drug and protects the health of patients. However, current observational research methods are subject to several biases. These include confounding by indication, reverse causality and missing data. We propose that Mendelian randomization (MR) offers a novel approach for the prediction of unintended drug effects. In particular, we advocate the synthesis of evidence from this method and other approaches, in the spirit of triangulation, to improve causal inferences concerning drug effects. MR addresses some of the limitations associated with the existing methods in this field. Furthermore, it can be applied either before or after approval of the drug, and could therefore prevent the potentially harmful exposure of patients in clinical trials and beyond. The potential of MR as a pharmacovigilance and drug repurposing tool is yet to be realized, and could both help prevent adverse drug events and identify novel indications for existing drugs in the future.

Elucidating the genetic basis of social interaction and isolation

The negative impacts of social isolation and loneliness on health are well documented. However, little is known about their possible biological determinants. In up to 452,302 UK Biobank study participants, we perform genome-wide association study analyses for loneliness and regular participation in social activities. We identify 15 genomic loci (P < 5 × 10⁻⁸) for loneliness, and demonstrate a likely causal association between adiposity and increased susceptibility to loneliness and depressive symptoms. Further loci were identified for regular attendance at a sports club or gym (N = 6 loci), pub or social club (N = 13) or religious group (N = 18). Across these traits there was strong enrichment for genes expressed in brain regions that control emotional expression and behaviour. We demonstrate aetiological mechanisms specific to each trait, in addition to identifying loci that are pleiotropic across multiple complex traits. Further study of these traits may identify novel modifiable risk factors associated with social withdrawal and isolation.

Little is known about the genetic determinants of social isolation and loneliness despite their well-established importance for health. Here, using multi-trait GWAS, Day et al. identify 15 genomic loci for loneliness and further show a bidirectional causal relationship between BMI and loneliness by MR.