INTRODUCTION:First-hand smoking is a major cause of global morbidity and mortality. Exposure to environmental tobacco smoke (ETS; "second-hand" or "passive smoking") may also cause ill health, but establishing ETS as the cause is challenging, in part due to confounding and reverse causation. METHODS:We applied Mendelian randomization (MR) to investigate the causal effects of ETS. We use four approaches to instrument ETS exposure: The first and second used an index individual's parent's genetically predicted smoking, independent of the index individual's genetically predicted smoking to assess the effects of that parent's smoking on the index individual. The third and fourth used one index individual's parent's genetically predicted smoking, independent of the other parent's genetically predicted smoking to assess the effects of the first parent's smoking on the second parent. We then meta-analyze the four MR approaches. RESULTS:Our findings suggest a causal effect of genetically predicted ETS exposure on lung cancer and chronic obstructive pulmonary disease (PFDR < .001 for both). We did not find evidence supporting an effect on hypertension, depression, coronary heart disease, or stroke (PFDR = 1.000 for all four non-respiratory outcomes). CONCLUSION:These results support existing public health measures to limit exposure to ETS. IMPLICATIONS:We assess the causal effects of environmental tobacco smoking (ETS; "second-hand smoking" or "passive smoking") using a quasi-experimental method, Mendelian randomization, which is more robust to confounding than conventional epidemiological methods.To study the effects of ETS exposure, we used an index individual's parent's or spouses' genetically predicted smoking, independent of either the index's or the other parent's genetically predicted smoking when assessing the effect of that parent's smoking on the index individual or other parent, respectively. We then meta-analyze the effects of different relatives.This study extends the Mendelian randomization paradigm to assess ETS by examining the effect that one relative has on the other relative, independent of the other relative's smoking. In doing so, it adds a unique source of evidence that triangulates with prior research to indicate an effect of ETS exposure on lung cancer and chronic obstructive pulmonary disease.

Target trial emulation prompts investigators to frame their analysis question in terms of a hypothetical clinical trial. Although this does not solve the problem of confounding, the framework can protect against other sources of bias. A natural question is this: what kinds of trials can be emulated? In a late-phase trial (Phase 3 or 4), the goal is to obtain a well-defined causal estimate that closely approximates the impact of a proposed intervention. In an early-phase trial (Phase 2 or earlier), the estimate is a means to an end rather than an end in itself. An early-phase trial provides proof-of-concept evidence on the impact of an intervention in the exposure in terms of efficacy and safety, but the estimand may not correspond to the intervention to be implemented in practice. In a natural experiment where causal inferences rely on a plausibly random (or quasi-random) comparison, the estimand may not be directly translatable to applied practice. In this case, the analysis may be conceptualized as an early-phase target trial. This provides less specific evidence than a late-stage target trial, but in many cases, a more valid but less applicable comparison is preferable to a more applicable comparison that is more susceptible to bias.

Abstract Confounding is a central challenge in observational studies. Here, we propose a framework for identifying confounders of two non-causally related traits by employing cross-trait pleiotropy analysis to detect genetic loci that affect both traits and multi-trait colocalisation to identify molecular phenotypes mediating these effects. We apply this approach to the analysis of C-reactive protein (CRP) - a non-specific marker of inflammation - and 10 inflammation-related cancers. In UK Biobank, higher pre-diagnostic CRP levels are associated with increased risk of multiple cancers, but bidirectional Mendelian randomization provides little evidence for a causal relationship. Cross-trait genetic analyses identify 92 loci with shared CRP-cancer effects including those with established roles in cancer and 50 novel loci such as RSPO3 (breast cancer) and GCKR (colorectal cancer). Integration with proteomic and single-cell transcriptomic data identified putative molecular mediators at 24 loci including plasma TLR1 levels in breast cancer and CD4 + T cell IRF5 expression in kidney cancer. Notably, 15 candidate effector genes encode targets of approved or investigational medications, including IL6 , PDE4D , and CASP8 , indicating potential opportunities for their repurposing for cancer prevention. The proposed approach provides a generalisable framework for leveraging non-causal phenotypic relationships to yield insights into disease mechanisms and therapeutic targets for disease prevention.

Cis-Mendelian randomization is a computational approach that uses genetic variants from a biologically relevant gene region to assess the plausibility of a causal effect of a specific mechanism on an outcome of interest. As the use of cis-Mendelian randomization has recently increased due to the abundance of genetic association data for molecular quantitative traits, there is a need for informed guidance on how best to conduct such studies. Here, we review and discuss the key considerations for conducting robust cis-Mendelian randomization analyses. The main considerations include selection of the gene region of interest, the choice of the best traits to proxy the exposure of interest, variant selection and validation, and relevant sensitivity analyses. We highlight the importance of incorporating biological insight throughout the whole analysis process and that the analytical methods should be tailored to each gene region. Moreover, we point out some key differences from genome-wide Mendelian randomization—where variants are selected across the genome—and emphasize that cis-Mendelian randomization requires a distinct set of sensitivity analyses. We believe the advice we provide in this review will lead to a higher standard in planning, conducting, reviewing, and interpreting cis-Mendelian randomization studies.

A longstanding aim of developmental psychology and epidemiology is to understand the causal effects of parental phenotypes on offspring outcomes. Traditional approaches often fail to account for confounding and reverse causation. We evaluate the use of Mendelian randomisation with non-inherited variants (MR-NIV) to address these limitations. MR-NIV leverages non-inherited genetic variants to instrument the parental phenotype independent of the offspring's genotype. We used Directed Acyclic Graphs and simulations to validate MR-NIV and explore robustness to assortative mating. In contrast to an alternative MR method which adjusts the parental genotype for offspring genotype, MR-NIV can be robust to assortative mating when used without trio data. In settings without trio data, MR-NIV outperformed the adjustment method. The adjustment method outperformed MR-NIV in settings with trio data. Applying MR-NIV to the Avon Longitudinal Study of Parents and Children, we assessed the causal effect of parental smoking on offspring smoking initiation at age 16. Results were consistent with observational studies, suggesting a meaningful increase in the risk of offspring smoking due to parental smoking. However, larger sample sizes will be necessary to provide a precise answer. MR-NIV offers a promising extension of Mendelian randomisation for studying the developmental environment.

Mendelian randomization (MR) makes causal claims by treating genetic variation in an analogous way to randomization in a clinical trial. MR investigations can be viewed as analogous to a randomized encouragement design, in that genetic variants do not determine the precise level of an exposure, but increase liability to it. As such, an MR estimate typically does not represent an achievable or well-defined causal effect in terms of the exposure, as it reflects the impact of a life-long shift in the trajectory of the exposure, which likely differs between individuals. We advocate for MR investigations to be performed to assess evidence for a causal hypothesis, rather than to estimate a well-defined causal quantity. MR estimates are useful to combine evidence across genetic variants, to assess the validity of variants as instruments, to provide confidence intervals, and to compare estimates across outcomes. However, numerical estimates from MR should not be over-interpreted. The value of an MR investigation is not to quantify the magnitude of effect for a well-defined intervention in the exposure. Instead, it provides a distinct source of evidence to increase or decrease confidence in a causal hypothesis, which can be triangulated with evidence from other sources.

Objectives:Evidence on the safety of TNF inhibitors (TNFi) for pregnancy-related maternal and foetal outcomes remains limited. While some studies report increased rates of preterm delivery, others have suggested a possible protective role for gestational diabetes. We used population-level data to examine the effect of genetically proxied TNFi on these outcomes. Methods:We proxied TNFi using rs1800693, a splicing variant within the TNFRSF1A gene, which is strongly associated with CRP in a genome-wide association study of 575 531 individuals of European ancestry. CRP was selected as the biomarker because TNFi is recognized to suppression CRP. Genetic association data for pregnancy-related outcomes were taken from FinnGen and UK Biobank, including the outcomes of spontaneous abortion, ectopic pregnancy, hyperemesis gravidarum, gestational diabetes, pre-eclampsia or eclampsia, preterm birth and offspring birthweight. Colocalization analysis was used to examine genetic confounding. Results:We found no strong association between genetically proxied TNFi and any adverse pregnancy-related outcome, including spontaneous abortion (odds ratio [OR] 1.07, 95% CI: 0.41, 2.81), preterm birth (OR 0.48, 95% CI 0.14, 1.60), hyperemesis gravidarum (OR 0.20, 95% CI 0.02, 2.57), pre-eclampsia or eclampsia (OR 0.59, 95% CI 0.13, 2.65). Genetically proxied TNFi was associated with lower risk of gestational diabetes (OR 0.16, 95% CI 0.05, 0.52). There was no statistical evidence to suggest genetic confounding through linkage disequilibrium. Conclusion:This genetic investigation found no evidence linking TNFi to adverse pregnancy-related outcomes. The suggestive association with a reduced risk of gestational diabetes warrants further research and may support its consideration for at-risk pregnant women.

Evans and colleagues for providing a critique on our article “Deriving GWAS summary estimates for paternal smoking in UK biobank: a GWAS by subtraction”. They highlight important limitations in our approach, which we overlooked. They argue that our approach works neither in theory nor in practice– that there are both flaws in the method and in our execution of the method. Here we explore these issues and address what we consider to be the most important criticisms.

While preclinical studies suggest that Phosphodiesterase 5 (PDE5) inhibition may reduce cognitive impairment, findings from observational studies on whether PDE5 inhibitors reduce Alzheimer's disease (AD) risk have been inconsistent. We performed a two-sample cis-Mendelian Randomisation (MR) analysis to estimate the causal effect of PDE5 inhibition on AD risk. The analysis was performed across four different genome-wide association studies (GWAS) of AD to enhance reliability through triangulation. Additionally, a sex-stratified MR analysis using data from UK Biobank was performed to assess potential sex-specific effects. No evidence of a causal association between PDE5 inhibition and AD risk was found in the main analyses. Similar findings were obtained in the sex-stratified analysis. Our study uses genetic data to triangulate the evidence and suggests that PDE5 inhibitors are unlikely to decrease the risk of AD. Further research is needed to thoroughly understand the impact of PDE5 inhibitors on the risk of Alzheimer's disease.

INTRODUCTION:Large observational and Mendelian randomization (MR) studies have demonstrated a strong association between both elevated LDL cholesterol (LDL-c) and triglycerides (TG) with risk of aortic stenosis (AS), although randomized trials showed no benefit of statins for AS. It consequently remains uncertain whether lipid-lowering therapies have a role to prevent or treat AS. We used a drug-target MR approach to investigate the genetically predicted effect of lipid-lowering therapies on risk of AS. METHODS AND RESULTS:We collected summary statistics for LDL-c, TG, and AS from genome-wide association studies (GWAS) including 1 320 016, 1 253 277, and 412 181 European participants from the Global Lipids Genetics Consortium and FinnGen study, respectively. We identified genetic proxies for PCSK9 inhibitors, statins, bempedoic acid, and ezetimibe as single nucleotide polymorphisms in or within 200 kb of the target genes (PCSK9, HMGCR, ACLY, and NPC1L1, respectively), which were also significantly associated with LDL-c at P < 5 × 10-8. We used a similar approach to identify genetic proxies for the TG-lowering agents fenofibrates, APOC3 inhibitors, and ANGPTL3 inhibitors using the target genes PPARA, APOC3, and ANGPTL3, respectively. Inverse variance-weighted was the primary analysis method. Sensitivity analyses included weighted median, weighted mode, and MR-Egger, followed by the outlier-exclusion approaches MR-PRESSO and Cook's distance. We also performed multivariable analyses to evaluate whether the predicted effect of PCSK9 inhibition may be mediated by lipoprotein(a). We performed replication and negative control analyses using GWAS of AS and height including 653 867 and 408 112 participants, respectively. Genetically proxied PCSK9 inhibition was significantly associated with reduced AS risk (odds ratio [OR] 0.61, 95% confidence interval [CI] 0.52-0.72, P < 0.0001) on main, replication, and all sensitivity analyses. Genetically proxied ezetimibe (OR 0.49, 95% CI 0.31-0.78, P = 0.003), bempedoic acid (OR 0.0054, 95% CI 0.0002-0.12, P = 0.0009), and statins (OR 0.61, 95% CI 0.46-0.81, P = 0.0006) were similarly associated with reduced AS risk, although the latter were not significant on replication analyses. Amongst the TG-lowering agents, genetically proxied APOC3 inhibition was associated with reduced AS risk (OR 0.78, 95% CI 0.70-0.88, P < 0.0001), but fenofibrate (OR 0.64, 95% CI 0.09-4.53, P = 0.65) and ANGPTL3 inhibitors (OR 1.05, 95% CI 0.77-1.43, P = 0.74) were not. CONCLUSIONS:Genetically proxied lipid-lowering therapies are significantly associated with reduced risk of AS. Early initiation and sustained administration of lipid-lowering therapies may prevent AS progression and warrants further research in the clinical trial setting.

Large observational and Mendelian randomisation (MR) studies have demonstrated a strong association between hypertension and risk of aortic stenosis (AS). Amongst the antihypertensive drugs, angiotensin converting enzyme (ACE) inhibitors are hypothesised to be most likely to demonstrate benefit for AS due to their antihypertensive and antifibrotic effects. We collected summary statistics for systolic blood pressure (SBP) and AS from genome-wide association studies including 1,028,980 and 956,682 European participants, respectively. We first performed conventional MR analyses to investigate a causal relationship between SBP and AS risk. Then drug-target MR was performed using distal and proximal effects approaches. In the distal effect drug-target MR analysis, we identified the genetic instrument for ACE inhibitors as single nucleotide polymorphisms within 200kb of the target ACE gene which were also significantly associated with SBP at p<5x10-8 and clumped using r2<0.01. In the proximal effect analysis we identified the genetic instrument for ACE inhibition as the lead SNP from expression quantitative trait locus analysis, then performed summary-data based Mendelian randomisation and heterogeneity in dependent instruments analysis (HEIDI). Genetically predicted SBP was significantly associated with increased risk of AS (OR 1.03 per mmHg increase in SBP, 95% CI 1.02-1.04, p<0.0001). On distal effect drug-target MR, genetically predicted ACE inhibition was significantly associated with reduced risk of AS (OR 0.92 per mmHg lower SBP, 95% CI 0.85-0.99, p=0.027). The validation analysis using the proximal effect drug-target MR approach demonstrated concordant results, with genetically predicted ACE inhibition significantly associated with reduced risk of AS (OR 0.69 per standard deviation decrease in gene expression, 95% 0.49-0.98, p=0.041) and the HEIDI analysis (p=0.57) was consistent with the presence of a shared causal variant. A positive control for the effect of ACE inhibition on risk of coronary artery disease demonstrated expected results (OR 0.90 per mmHg lower SBP, 95% CI 0.85-0.96, p=0.0004). This study provides genetic evidence supporting a potential therapeutic role for ACE inhibitors in reducing the risk of AS. The concordance between both proximal (gene expression) and distal (blood pressure) genetic approaches strengthens the validity of the findings, and provides a strong rationale for randomised controlled trials of ACE inhibitors for AS.

Many Mendelian randomization (MR) papers have been conducted only in people of European ancestry, limiting transportability of results to the global population. Expanding MR to diverse ancestry groups is essential to ensure equitable biomedical insights, yet presents analytical and conceptual challenges. This review examines the practical challenges of MR analyses beyond the European only context, including use of data from multi-ancestry, mismatched ancestry, and admixed populations. We explain how apparent heterogeneity in MR estimates between populations can arise from differences in genetic variant frequencies and correlation patterns, as well as from differences in the distribution of phenotypic variables, complicating the detection of true differences in the causal pathway. We summarize published strategies for selecting genetic instruments and performing analyses when working with limited ancestry-specific data, discussing the assumptions needed in each case for incorporating external data from different ancestry populations. We conclude that differences in MR estimates by ancestry group should be interpreted cautiously, with consideration of how the identified differences may arise due to social and cultural factors. Corroborating evidence of a biological mechanism altering the causal pathway is needed to support a conclusion of differing causal pathways between ancestry groups.

Higher consumption of caffeinated beverages is associated with disturbed sleep patterns. Using genetic variants as proxies for caffeine consumption, caffeine metabolism, and sleep traits, we investigated whether this association reflects a direct effect of caffeine. Genetic variants associated with caffeine consumption (n = 407,072), caffeine metabolism (n = 9876), chronotype (n = 449,734), daytime napping (n = 452,633), daytime sleepiness (n = 452,071), getting up in morning (n = 385,949), insomnia (n = 453,379), and sleep duration (n = 446,118) identified in individuals from several studies, including the UK Biobank, were used to explore bi-directional causal relationships between caffeine and sleep using a series of univariable Mendelian Randomisation analyses. We used multivariable Mendelian Randomisation to explore the direct effects of caffeine consumption on sleep behaviours while adjusting for metabolism and vice versa. Higher consumption decreased daytime sleepiness (β univariable = -0.044, 95% CI [-0.065, -0.023], p < 0.001; β multivariable = -0.034, 95% CI [-0.058, -0.009], p = 0.010), while faster caffeine metabolism, indicative of less caffeine exposure per beverage consumed, decreased the likelihood of daytime napping (β univariable = -0.024, 95% CI [-0.037, -0.011], p < 0.001; β multivariable = -0.021, 95% CI [-0.042, 0.000], p = 0.051). Being an evening person decreased caffeine consumption (β univariable = -0.044, 95% CI [-0.078, -0.010], p = 0.010). Caffeine consumption/metabolism was not causally related to sleep duration or insomnia. We found no clear evidence for effects of caffeine consumption/metabolism on sleep among non-current caffeine consumers when assessing possible pleiotropy. Overall, sleep appears to be impacted by caffeine in a way that influences daytime alertness rather than night-time sleep characteristics. However, the presence of weak instruments for caffeine metabolism and significant heterogeneity warrants further research with larger and diverse samples to better understand the causal pathway between caffeine and sleep.

Smoking is a major cause of global morbidity and premature mortality. Exposure to environmental tobacco smoke (ETS; “second-hand” or “passive smoking”) may also contribute to ill health. However, it is difficult to establish causality given problems of confounding and reverse causation. We applied Mendelian randomisation to investigate evidence for causal effects. To instrument ETS exposure we used an index individual’s parent’s or spouse’s genetic liability to smoke, conditional on the index individual’s genetic liability. We then meta-analyse four MR approaches using this. Our findings suggest a causal effect of genetically predicted ETS exposure on lung cancer and chronic obstructive pulmonary disease (pFDR < 0.001 for both). We did not find evidence supporting an effect on hypertension, depression, coronary heart disease, or stroke (pFDR = 1.000 for all four non-respiratory outcomes); but this might reflect low statistical power. Overall, these results support public health measures to limit exposure to ETS.

Background Mendelian randomisation (MR) is the use of genetic variants as instrumental variables. Mode-based estimators (MBE) are one of the most popular types of estimators used in univariable-MR studies and is often used as a sensitivity analysis for pleiotropy. However, because there are no plurality valid regression estimators, modal estimators for multivariable-MR have been under-explored. Methods We use the residual framework for multivariable-MR to introduce two multivariable modal estimators: multivariable-MBE, which uses IVW to create residuals fed into a traditional plurality valid estimator, and an estimator which instead has the residuals fed into the contamination mixture method (CM), multivariable-CM. We then use Monte-Carlo simulations to explore the performance of these estimators when compared to existing ones and re-analyse the data used by Grant and Burgess (2021) looking at the causal effect of intelligence, education, and household income on Alzheimer’s disease as an applied example. Results In our simulation, we found that multivariable-MBE was generally too variable to be much use. Multivariable-CM produced more precise estimates on the other hand. Multivariable-CM performed better than MR-Egger in almost all settings, and Weighted Median under balanced pleiotropy. However, it underperformed Weighted Median when there was a moderate amount of directional pleiotropy. Our re-analysis supported the conclusion of Grant and Burgess (2021), that intelligence had a protective effect on Alzheimer’s disease, while education, and household income do not have a causal effect. Conclusions Here we introduced two, non-regression-based, plurality valid estimators for multivariable MR. Of these, “multivariable-CM” which uses IVW to create residuals fed into a contamination-mixture model, performed the best. This estimator uses a plurality of variants valid assumption, and appears to provide precise and unbiased estimates in the presence of balanced pleiotropy and small amounts of directional pleiotropy.

Abstract Background Genome-wide association studies have enabled Mendelian randomization analyses to be performed at an industrial scale. Two-sample summary data Mendelian randomization analyses can be performed using publicly available data by anyone who has access to the internet. While this has led to many insightful papers, it has also fuelled an explosion of poor-quality Mendelian randomization publications, which threatens to undermine the credibility of the whole approach. Findings We detail five pitfalls in conducting a reliable Mendelian randomization investigation: (1) inappropriate research question, (2) inappropriate choice of variants as instruments, (3) insufficient interrogation of findings, (4) inappropriate interpretation of findings, and (5) lack of engagement with previous work. We have provided a brief checklist of key points to consider when performing a Mendelian randomization investigation; this does not replace previous guidance, but highlights critical analysis choices. Journal editors should be able to identify many low-quality submissions and reject papers without requiring peer review. Peer reviewers should focus initially on key indicators of validity; if a paper does not satisfy these, then the paper may be meaningless even if it is technically flawless. Conclusions Performing an informative Mendelian randomization investigation requires critical thought and collaboration between different specialties and fields of research.

Background Caffeine is one of the most utilized drugs in the world, yet its clinical effects are not fully understood. Circulating caffeine levels are influenced by the interplay between consumption behaviour and metabolism. This study aimed to investigate the effects of circulating caffeine levels by considering genetically predicted variation in caffeine metabolism. Methods Leveraging genetic variants related to caffeine metabolism that affect its circulating levels, we investigated the clinical effects of plasma caffeine in a phenome-wide association study (PheWAS). We validated novel findings using a two-sample Mendelian randomization framework and explored the potential mechanisms underlying these effects in proteome-wide and metabolome-wide Mendelian randomization. Results Higher levels of genetically predicted circulating caffeine among caffeine consumers were associated with a lower risk of obesity (odds ratio (OR) per standard deviation increase in caffeine = 0.97, 95% confidence interval (CI) CI: 0.95—0.98, p = 2.47 × 10 −4 ), osteoarthrosis (OR = 0.97, 95% CI: 0.96—0.98, P=1.10 × 10 −8 ) and osteoarthritis (OR: 0.97, 95% CI: 0.96 to 0.98, P = 1.09 × 10 −6 ). Approximately one third of the protective effect of plasma caffeine on osteoarthritis risk was estimated to be mediated through lower bodyweight. Proteomic and metabolomic perturbations indicated lower chronic inflammation, improved lipid profiles, and altered protein and glycogen metabolism as potential biological mechanisms underlying these effects. Conclusions We report novel evidence suggesting that long-term increases in circulating caffeine may reduce bodyweight and the risk of osteoarthrosis and osteoarthritis. We confirm prior genetic evidence of a protective effect of plasma caffeine on risk of overweight and obesity. Further clinical study is warranted to understand the translational relevance of these findings before clinical practice or lifestyle interventions related to caffeine consumption are introduced.

Two-sample MR is an increasingly popular method for strengthening causal inference in epidemiological studies. For the effect estimates to be meaningful, variant-exposure and variant-outcome associations must come from comparable populations. A recent systematic review of two-sample MR studies found that, if assessed at all, MR studies evaluated this assumption by checking that the genetic association studies had similar demographics. However, it is unclear if this is sufficient because less easily accessible factors may also be important. Here we propose an easy-to-implement falsification test. Since recent theoretical developments in causal inference suggest that a causal effect estimate can generalise from one study to another if there is exchangeability of effect modifiers, we suggest testing the homogeneity of variant-phenotype associations for a phenotype which has been measured in both genetic association studies as a method of exploring the ‘same-population’ test. This test could be used to facilitate designing MR studies with diverse populations. We developed a simple R package to facilitate the implementation of our proposed test. We hope that this research note will result in increased attention to the same-population assumption, and the development of better sensitivity analyses.