Mendelian Randomization in Cardiovascular Research: Establishing Causality When There Are Unmeasured Confounders

The Causal Relationship between PCSK9 Inhibitors and Osteoporosis Based on Drug-Targeted Mendelian Combined Mediation Analysis

PCSK9 inhibitors have been shown to lower serum low density lipoprotein cholesterol (LDL-C) levels and are considered integral in the treatment of cardiovascular diseases. However, the potential association between PCSK9 inhibitors and osteoporosis is unclear now. In this study, drug-targeted mendelian randomization (MR) was utilized in conjunction with mediation analysis including bone mineral density (BMD), total 25-hydroxyvitamin D (T25(OH)D) levels and calcium supplementation to investigate the causal relationship between PCSK9 inhibitors and osteoporosis. The LDL-C level was chosen as the exposure variable in a sample size of 173,082 individuals. We conducted a MR analysis on the relationship between PCSK9 inhibitors and osteoporosis, elucidating the mediators involved. Utilizing the inverse variance weighted (IVW) method, we found the risk of osteoporosis was reduced by 0.6% in those who used PCSK9 inhibitors compared with non-users (OR: 0.994, 95%CI: 0.991-0.998, P < 0.001). In people aged 30-45 years, the risk of low BMD was 1.176 times higher among PCSK9 inhibitor users compared to non-users (OR: 1.176, 95%CI: 1.017-1.336, P = 0.045). Conversely, people aged 45-60 years who used PCSK9 inhibitors had a 14.9% lower risk of low BMD compared to non-users (OR: 0.851, 95%CI: 0.732-0.968, P = 0.007). Mediation analysis revealed that 43.33% of the impact of PCSK9 inhibitors on osteoporosis was mediated through BMD levels, with the remaining 56.67% being a direct effect. Effects of PCSK9 inhibitors on BMD levels varied in different ages. In addition, the risk of high serum T25(OH)D levels were 1.091 times among PCSK9 inhibitor users compared to non-users (OR: 1.091, 95%CI: 1.065-1.112, P < 0.001), providing valuable insights for clinicians.

Exploring machine learning strategies for predicting cardiovascular disease risk factors from multi-omic data

BACKGROUND

Machine learning (ML) classifiers are increasingly used for predicting cardiovascular disease (CVD) and related risk factors using omics data, although these outcomes often exhibit categorical nature and class imbalances. However, little is known about which ML classifier, omics data, or upstream dimension reduction strategy has the strongest influence on prediction quality in such settings. Our study aimed to illustrate and compare different machine learning strategies to predict CVD risk factors under different scenarios.

METHODS

We compared the use of six ML classifiers in predicting CVD risk factors using blood-derived metabolomics, epigenetics and transcriptomics data. Upstream omic dimension reduction was performed using either unsupervised or semi-supervised autoencoders, whose downstream ML classifier performance we compared. CVD risk factors included systolic and diastolic blood pressure measurements and ultrasound-based biomarkers of left ventricular diastolic dysfunction (LVDD; E/e' ratio, E/A ratio, LAVI) collected from 1,249 Finnish participants, of which 80% were used for model fitting. We predicted individuals with low, high or average levels of CVD risk factors, the latter class being the most common. We constructed multi-omic predictions using a meta-learner that weighted single-omic predictions. Model performance comparisons were based on the F1 score. Finally, we investigated whether learned omic representations from pre-trained semi-supervised autoencoders could improve outcome prediction in an external cohort using transfer learning.

RESULTS

Depending on the ML classifier or omic used, the quality of single-omic predictions varied. Multi-omics predictions outperformed single-omics predictions in most cases, particularly in the prediction of individuals with high or low CVD risk factor levels. Semi-supervised autoencoders improved downstream predictions compared to the use of unsupervised autoencoders. In addition, median gains in Area Under the Curve by transfer learning compared to modelling from scratch ranged from 0.09 to 0.14 and 0.07 to 0.11 units for transcriptomic and metabolomic data, respectively.

CONCLUSIONS

By illustrating the use of different machine learning strategies in different scenarios, our study provides a platform for researchers to evaluate how the choice of omics, ML classifiers, and dimension reduction can influence the quality of CVD risk factor predictions.

Genetically Predicted Apolipoprotein E Levels with the Risk of Panvascular Diseases: A Mendelian Randomization Study

The aim of this study was to comprehensively assess the causal relationship between the overall genetic effect of circulating ApoE levels and panvascular lesions using newer genome-wide association data and two-sample bidirectional Mendelian randomization (MR) analysis. Two-way MR using single-nucleotide polymorphisms of circulating ApoE as instrumental variables was performed using the highest-priority Genome-wide association study (GWAS) data, with factor-adjusted and data-corrected statistics, to estimate causal associations between circulating ApoE levels and 10 pan-vascular diseases in > 500,000 UK Biobank participants, > 400,000 participants of Finnish ancestry, and numerous participants in a consortium of predominantly European ancestry. Meta-analysis was conducted to assess positive results. After correcting for statistical results, elevated circulating ApoE levels were shown to have a significant protective effect against Cerebral ischemia (CI) [IVW odds ratio (OR) 0.888, 95% Confidence Interval (CI): 0.823-0.958, p = 2.3 × 10-3], Coronary heart disease [IVW OR 0.950,95% CI: 0.924-0.976, p = 2.0 × 10-4] had a significant protective effect and potentially suggestive protective causality against Angina pectoris [IVW odds ratio (OR) 0.961, 95%CI: 0.931-0.991, p = 1.1 × 10-2]. There was a potential causal effect for increased risk of Heart failure (HF) [IVW ratio (OR) 1.040, 95%CI: 1.006-1.060, p = 1.8 × 10-2]. (Bonferroni threshold p < 0.0026, PFDR < 0.05) Reverse MR analysis did not reveal significant evidence of a causal effect of PVD on changes in circulating ApoE levels. Meta-analysis increases reliability of results. Elevated circulating ApoE levels were particularly associated with an increased risk of heart failure. Elevated ApoE levels reduce the risk of cerebral ischemia, coronary heart disease, and angina pectoris, reflecting a protective effect. The possible pathophysiological role of circulating ApoE levels in the development of panvascular disease is emphasized.

Associations between gut microbiota and gynecological cancers: A bi-directional two-sample Mendelian randomization study

Growing evidence has suggested that gut microbiota is associated with gynecologic cancers. However, whether there is a causal relationship between these associations remains to be determined. A two-sample Mendelian randomization (MR) evaluation was carried out to investigate the mechanism associating gut microbiota and 3 prevalent gynecological cancers, ovarian cancer (OC), endometrial cancer, and cervical cancer as well as their subtypes in individuals of European ancestry. The Genome-wide association studies statistics, which are publically accessible, were used. Eligible instrumental single nucleotide polymorphisms that were significantly related to the gut microbiota were selected. Multiple MR analysis approaches were carried out, including inverse variance weighted, MR-Egger, Weighted Median methods, and a range of sensitivity analyses. Lastly, we undertook a reverse MR analysis to evaluate the potential of reverse causality. We sifted through 196 bacterial taxa and identified 33 suggestive causal relationships between genetic liability in the gut microbiota and gynecological cancers. We found that 11 of these genera could be pathogenic risk factors for gynecological cancers, while 19 could lessen the risk of cancer. In the other direction, gynecological cancers altered gut microbiota composition. Our MR analysis revealed that the gut microbiota was causally associated with OC, endometrial cancer, and cervical cancer. This may assist in providing new insights for further mechanistic and clinical studies of microbiota-mediated gynecological cancer.

Deciphering thyroid function and CIMT: a Mendelian randomization study of the U-shaped influence mediated by apolipoproteins

Background

Carotid Intima-Media Thickness (CIMT) is a key marker for atherosclerosis, with its modulation being crucial for cardiovascular disease (CVD) risk assessment. While thyroid function's impact on cardiovascular health is recognized, the causal relationship and underlying mechanisms influencing CIMT remain to be elucidated.

Methods

In this study, Mendelian Randomization (MR) was employed to assess the causal relationship between thyroid function and CIMT. Thyroid hormone data were sourced from the Thyroidomics Consortium, while lipid traits and CIMT measurements were obtained from the UK Biobank. The primary analysis method was a two-sample MR using multiplicative random effects inverse variance weighting (IVW-MRE). Additionally, the study explored the influence of thyroid hormones on lipid profiles and assessed their potential mediating role in the thyroid function-CIMT relationship through multivariate MR analysis.

Results

The study revealed that lower levels of Free Thyroxine (FT4) within the normal range are significantly associated with increased CIMT. This association was not observed with free triiodothyronine (FT3), thyroid-stimulating hormone (TSH), or TPOAb. Additionally, mediation analysis suggested that apolipoprotein A-I and B are involved in the relationship between thyroid function and CIMT. The findings indicate a potential U-shaped curve relationship between FT4 levels and CIMT, with thyroid hormone supplementation in hypothyroid patients showing benefits in reducing CIMT.

Conclusion

This research establishes a causal link between thyroid function and CIMT using MR methods, underscoring the importance of monitoring thyroid function for early cardiovascular risk assessment. The results advocate for the consideration of thyroid hormone supplementation in hypothyroid patients as a strategy to mitigate the risk of carotid atherosclerosis. These insights pave the way for more targeted approaches in managing patients with thyroid dysfunction to prevent cardiovascular complications.

Causal associations between common musculoskeletal disorders and dementia: a Mendelian randomization study

Introduction

Dementia and musculoskeletal disorders (MSDs) are major public health problems. We aimed to investigate the genetic causality of common MSDs and dementia.

Methods

Two-sample Mendelian randomization (MR) was used in this study. MR analysis based on gene-wide association study (GWAS) data on osteoarthritis (OA), dementia with Lewy bodies, and other MSDs and dementia types were obtained from the Genetics of Osteoarthritis consortium, IEU-open GWAS project, GWAS catalog, and FinnGen consortium. Rigorously selected single-nucleotide polymorphisms were regarded as instrumental variables for further MR analysis. Inverse-variance weighted, MR-Egger regression, weight median, simple mode, and weight mode methods were used to obtain the MR estimates. Cochran's Q test, MR-Egger and MR-Pleiotropy Residual Sum and Outlier analysis, and the leave-one-out test were applied for sensitivity testing.

Results

The inverse-variance weighted method showed that hip OA was genetically associated with a lower risk of dementia, unspecified dementia, dementia in Alzheimer's disease, and vascular dementia. Kneehip OA was inversely associated with unspecified dementia and vascular dementia. Rheumatoid arthritis, juvenile idiopathic arthritis and seronegative rheumatoid arthritis were inversely associated with frontotemporal dementia, and rheumatoid arthritis was inversely associated with unspecified dementia. Simultaneously, ankylosing spondylitis was an independent risk factor for dementia, dementia with Lewy bodies, and dementia in Alzheimer's disease. Sensitivity tests showed that heterogeneity and horizontal pleiotropy did not exist in these associations. The leave-one-out test showed that these associations were stable.

Conclusion

We found that some MSDs were associated with the risk of dementia and provide evidence for the early detection of dementia in patients with MSDs and for the impact of inflammation on the central nervous system.

Genomic variation associated with cardiovascular disease progression following preeclampsia: a systematic review

Background

Women with a history of preeclampsia (PE) have been shown to have up to five times the risk of developing later-life cardiovascular disease (CVD). While PE and CVD are known to share clinical and molecular characteristics, there are limited studies investigating their shared genomics (genetics, epigenetics or transcriptomics) variation over time. Therefore, we sought to systematically review the literature to identify longitudinal studies focused on the genomic progression to CVD following PE.

Methods

A literature search of primary sources through PubMed, Scopus, Web of Science and Embase via OVID was performed. Studies published from January 1, 1980, to July 28, 2023, that investigated genomics in PE and CVD were eligible for inclusion. Included studies were screened based on Cochrane systematic review guidelines in conjunction with the PRISMA 2020 checklist. Eligible articles were further assessed for quality using the Newcastle-Ottawa scale.

Results

A total of 9,231 articles were screened, with 14 studies subjected to quality assessment. Following further evaluation, six studies were included for the final review. All six of these studies were heterogeneous in regard to CVD/risk factor as outcome, gene mapping approach, and in different targeted genes. The associated genes were RGS2, LPA, and AQP3, alongside microRNAs miR-122-5p, miR-126-3p, miR-146a-5p, and miR-206. Additionally, 12 differentially methylated regions potentially linked to later-life CVD following PE were identified. The only common variable across all six studies was the use of a case-control study design.

Conclusions

Our results provide critical insight into the heterogeneous nature of genomic studies investigating CVD following PE and highlight the urgent need for longitudinal studies to further investigate the genetic variation underlying the progression to CVD following PE.

Chromosome Y Haplogroup R Was Associated with the Risk of Premature Myocardial Infarction with ST-Elevation: Data from the CholeSTEMI Registry

Cardiovascular disease (CVD) is the leading cause of death worldwide, with coronary artery disease (CAD) being one of its main manifestations. Both environmental and genetic factors are widely known to be related to CAD, such as smoking, diabetes mellitus, dyslipidemia, and a family history of CAD. However, there is still a lack of information about other risk factors, especially those related to genetic mutations. Sex represents a classic CAD risk factor, as men are more likely to suffer CAD, but there is lack of evidence with regard to sex-specific genetic factors. We evaluated the Y chromosome haplogroups in a cohort of young Spanish male patients who suffered from STEMI. In this cohort, haplogroup R was significantly more frequent in STEMI patients.

Air Pollution, Built Environment, and Early Cardiovascular Disease

As the world's population becomes increasingly urbanized, there is growing concern about the impact of urban environments on cardiovascular health. Urban residents are exposed to a variety of adverse environmental exposures throughout their lives, including air pollution, built environment, and lack of green space, which may contribute to the development of early cardiovascular disease and related risk factors. While epidemiological studies have examined the role of a few environmental factors with early cardiovascular disease, the relationship with the broader environment remains poorly defined. In this article, we provide a brief overview of studies that have examined the impact of the environment including the built physical environment, discuss current challenges in the field, and suggest potential directions for future research. Additionally, we highlight the clinical implications of these findings and propose multilevel interventions to promote cardiovascular health among children and young adults.

Circulating Soluble CD163, Associations With Cardiovascular Outcomes and Mortality, and Identification of Genetic Variants in Older Individuals: The Cardiovascular Health Study

Background Monocytes/macrophages participate in cardiovascular disease. CD163 (cluster of differentiation 163) is a monocyte/macrophage receptor, and the shed sCD163 (soluble CD163) reflects monocyte/macrophage activation. We examined the association of sCD163 with incident cardiovascular disease events and performed a genome-wide association study to identify sCD163-associated variants. Methods and Results We measured plasma sCD163 in 5214 adults (aged ≥65 years, 58.7% women, 16.2% Black) of the CHS (Cardiovascular Health Study). We used Cox regression models (associations of sCD163 with incident events and mortality); median follow-up was 26 years. Genome-wide association study analyses were stratified on race. Adjusted for age, sex, and race and ethnicity, sCD163 levels were associated with all-cause mortality (hazard ratio [HR], 1.08 [95% CI, 1.04-1.12] per SD increase), cardiovascular disease mortality (HR, 1.15 [95% CI, 1.09-1.21]), incident coronary heart disease (HR, 1.10 [95% CI, 1.04-1.16]), and incident heart failure (HR, 1.18 [95% CI, 1.12-1.25]). When further adjusted (eg, cardiovascular disease risk factors), only incident coronary heart disease lost significance. In European American individuals, genome-wide association studies identified 38 variants on chromosome 2 near MGAT5 (top result rs62165726, P=3.3×10^-18),19 variants near chromosome 17 gene ASGR1 (rs55714927, P=1.5×10^-14), and 18 variants near chromosome 11 gene ST3GAL4. These regions replicated in the European ancestry ADDITION-PRO cohort, a longitudinal cohort study nested in the Danish arm of the Anglo-Danish-Dutch study of Intensive Treatment Intensive Treatment In peOple with screeNdetcted Diabetes in Primary Care. In Black individuals, we identified 9 variants on chromosome 6 (rs3129781 P=7.1×10^-9) in the HLA region, and 3 variants (rs115391969 P=4.3×10^-8) near the chromosome 16 gene MYLK3_. Conclusions Monocyte function, as measured by sCD163, may be predictive of overall and cardiovascular-specific mortality and incident heart failure.

Recommendations for Statistical Reporting in Cardiovascular Medicine: A Special Report From the American Heart Association

Statistical analyses are a crucial component of the biomedical research process and are necessary to draw inferences from biomedical research data. The application of sound statistical methodology is a prerequisite for publication in the American Heart Association (AHA) journal portfolio. The objective of this document is to summarize key aspects of statistical reporting that might be most relevant to the authors, reviewers, and readership of AHA journals. The AHA Scientific Publication Committee convened a task force to inventory existing statistical standards for publication in biomedical journals and to identify approaches suitable for the AHA journal portfolio. The experts on the task force were selected by the AHA Scientific Publication Committee, who identified 12 key topics that serve as the section headers for this document. For each topic, the members of the writing group identified relevant references and evaluated them as a resource to make the standards summarized herein. Each section was independently reviewed by an expert reviewer who was not part of the task force. Expert reviewers were also permitted to comment on other sections if they chose. Differences of opinion were adjudicated by consensus. The standards presented in this report are intended to serve as a guide for high-quality reporting of statistical analyses methods and results.