Causal Characteristics of Immune Cells Associated with Aortic Dissection: A Mendelian Randomisation Analysis

The diagnostic and therapeutic landscape for aortic dissection (AD) involves a tapestry of intricate clinical evaluations, sophisticated imaging modalities and a holistic assessment of patient health dynamics.^1,2 Within this framework, the immunological profile has surfaced as a pivotal determinant in the pathogenesis of AD, with particular immune cell phenotypes being implicated in its evolution.^3,4 Yet, the delineation of a definitive causal nexus between immune cell behaviour and AD continues to be elusive.

Mendelian randomisation (MR) has emerged as a pivotal methodology in the field of epidemiological research, harnessing genetic variants as proxies for environmental exposures to bolster causal assertions.⁵ The inherent randomness with which these genetic markers are distributed at conception mitigates the potential for residual confounding, diminishing the likelihood of entanglement with environmental or behavioural influences.⁶ This aspect of MR is particularly advantageous as it avoids the pitfalls of reverse causation, given that genetic predispositions are established prior to the manifestation of disease and remain unaltered by its progression. Furthermore, these genetic markers encapsulate the cumulative exposure over an individual’s lifetime, thereby minimising the impact of regression dilution bias.⁷

Building on these methodological strengths, we conducted a two-sample MR analysis to explore the immunological factors contributing to AD. This research aims to revolutionise therapeutic approaches and enhance clinical management of this critical vascular condition.

Methods

Research Design

Our investigation adopted a bivariate MR framework to probe the causal dynamics between an array of 731 immune cell attributes and AD. MR leverages genetic variants as instrumental variables (IVs) for risk factors, which are required to meet three core conditions to be considered valid causes: genetic instruments must demonstrate a robust association with the exposure of interest; they must be insulated from confounding variables; and their impact on the outcomes should be mediated solely via their association with the exposure.^6,8

Data Compilation

AD data, consisting of 967 cases and 381,977 controls, were obtained from the FinnGen Consortium (https://r6.finngen.fi).⁹ For the immune cell features, we used publicly accessible genome-wide association study (GWAS) summary statistics for 731 immune phenotypes from the GWAS Catalogue (IDs GCST0001391 to GCST0002121) (https://www.ebi.ac.uk/gwas/).¹⁰ The dataset included absolute cell counts (AC) for 118 phenotypes, median fluorescence intensity (MFI) for 389 phenotypes indicating surface antigen levels, morphological parameters (MP) for 32 phenotypes and relative cell counts (RC) for 192 phenotypes. These encompassed various immune cell types, including B cells, classic dendritic cells, T cell maturation stages, monocytes, myeloid cells and the T cells, B cells, and natural killer cells (TBNK) and regulatory T cells (Treg) panels. The GWAS for these immune traits was conducted on a European cohort of 3,757 individuals, with covariates such as sex, age and approximately 22 million single nucleotide polymorphisms (SNPs) genotyped and imputed using high-density arrays and a European reference panel.

Ethics Statement

This study was carried out in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki). The data used in this study were obtained from publicly accessible databases, specifically the FinnGen Consortium and the GWAS Catalogue. These datasets have already received approval from their respective ethics committees, and written informed consent was obtained from all participants in the original studies.

Single Nucleotide Polymorphisms Quality Assurance

To maintain analytical integrity, we instituted a rigorous SNP quality control protocol. This included: exclusion of non-biallelic SNPs; elimination of SNPs with ambiguous alleles (A/T, C/G); removal of SNPs lacking rsID numbers, with duplicate rsIDs, or without precise base pair (BP) positioning; discarding SNPs absent from the 1,000 Genomes Project Phase 3 or equivalent reference panels; rejection of SNPs with mismatched BP positions or alleles compared to the reference panel; exclusion of SNPs with an imputation INFO score below 0.9 which indicates it is imperfect; elimination of SNPs located on the X and Y chromosomes and mitochondria; removal of SNPs with sample sizes deviating by more than five SD from the mean.^11,12 This ensured the exclusion of SNPs that were genotyped on specialised arrays with disproportionately large sample sizes.

Instrumental Variable Selection

In our MR analysis, different significance thresholds were applied for selecting SNPs due to the varying sample sizes between immune cell traits and AD. For immune cell traits, we set a significance threshold of p<1×10⁻⁵ to increase the number of instrumental variables. This threshold was chosen because the sample size for immune cell traits was relatively small (n=3,757), necessitating a more lenient threshold to ensure sufficient statistical power.¹³ Conversely, we employed a stricter genome-wide significance threshold of p<5×10⁻⁸ for AD due to the much larger sample size (n=382,944).⁹ This larger sample size provided a greater power to detect robust associations, justifying the use of a more stringent threshold to maintain specificity. We used the clumping feature, which allows grouping of statistically significant genetic variants that are located in close proximity and are in linkage disequilibrium (LD) in PLINK – a free tool for analysing genotype and phenotype data – to remove dependent instrumental variables with an r²<0.001, using the European 1,000 Genomes Project panel as a reference.¹⁴ When an instrumental variable was missing in the outcome data, we replaced it with a proxy that had an r²>0.8.¹⁵ The strength of each instrument was assessed using the F statistic. We calculated the r² value to determine how much each genetic variant explained the variability in the exposure.¹⁶ A suite of visual aids, including scatter plots, funnel plots, and leave-one-out cross-validation (LOOCV), elucidated the genetic variants’ effects, validating the IVs’ reliability and coherence. The specific SNPs used in our study are available on request.

Mendelian Randomisation Execution

During the univariable MR analysis phase, we quantified the cumulative effect of risk factors on outcome probability through a two-sample MR analysis employing the fixed/random-effects inverse-variance weighting (IVW) method as our principal analytic approach. The association between each exposure and outcome was represented as an OR per unit increase in the genetically predicted exposure trait. This method is well known for its effectiveness and efficiency when all necessary conditions are met. However, deviations from the assumptions of the IVW method can lead to biased MR results. We applied a Bonferroni-corrected significance threshold of 0.005 to account for multiple comparisons in our primary analysis, while a p-value between 0.05 and 0.005 was considered suggestive.

To ensure the robustness of our findings, we conducted sensitivity analyses using a range of MR methodologies, each supported by unique assumptions and advantages. The weighted median approach offers a robust causal estimate, assuming valid genetic variants contribute the majority of the analytical weight.¹⁷ MR estimates are prioritised by their weighted significance, with the median value representing the consensus MR estimate. MR-Egger regression evaluates variant-outcome associations against variant-exposure correlations, weighted by the precision of the variant-outcome estimates.¹⁸ This technique detects horizontal pleiotropy and provides a valid MR estimate under the instrument strength independent of direct effect (InSIDE) assumption, applicable when variant pleiotropic effects on the outcome are independent of their exposure associations. Additionally, we used directional inverse-variance weighting (D-IVW) and the Mendelian randomisation robust adjusted profile score (MR-RAPS). D-IVW, a modification of IVW, accounts for directional pleiotropy and offers more precise causal estimates in the presence of suspected pleiotropy.¹⁹ MR-RAPS enhances robustness by adjusting for weak instruments and residual pleiotropy, ensuring more reliable inferences when standard MR assumptions are violated.²⁰

Horizontal pleiotropy, which occurs when a genetic variant influences the outcome independently of the exposure pathway, can lead to spurious causal inferences in MR studies.²⁰ This phenomenon is categorised into uncorrelated pleiotropy, where distinct mechanisms drive the genetic variant’s effect on exposure and outcome, and correlated pleiotropy, where a genetic variant affects both through a shared pathway.²¹ The MR-Egger intercept test and Mendelian Randomisation Pleiotropy RESidual Sum and Outlier (MR-PRESSO) were then used to investigate horizontal pleiotropy, confirming the genetic variants’ separate impact on both exposure and outcome. The MR-PRESSO framework identifies potential horizontal pleiotropy by regressing the effects of genetic variants on the outcome against their impact on the exposure, excluding outlier variants to correct for pleiotropy, and subsequently reapplying the IVW method without these outliers.¹⁷

These methodologies were implemented using various R packages designed for MR analysis, ensuring a rigorous and transparent evaluation of the causal relationships under investigation. Moreover, if multiple immune cell features were implicated, our multivariable MR analysis would concurrently address the potential horizontal pleiotropy across these traits. Reverse MR analysis mitigated concerns regarding reverse causation.

Pathway and Tissue Enrichment Exploration

We pinpointed significant cis-expression quantitative trait loci (cis-eQTLs) via the Genotype-Tissue Expression (GTEx) Consortium database, uncovering variations in gene expression. We mapped risk loci genes associated with causative immune cells identified at a false discovery rate (FDR) threshold of FDR<0.10. Subsequent tissue-specific expression and pathway analyses were performed on these pivotal genes using functional mapping and annotation for tissue-specific expression and enrichment analyses. Additionally, we used Metascape tools and the clusterProfiler package in R software to integrate Gene Ontology (GO), Kyoto Encyclopaedia of Genes and Genomes (KEGG), WikiPathways, and cell-specific gene set databases for extensive pathway analysis.^22,23

Friends Analysis

The Friends analysis is a functional clustering analysis of gene sets. By constructing a gene interaction network and using network topology parameters, the importance of each gene is calculated. This method allows for the analysis and prediction of the functional roles and regulatory mechanisms of different genes in related biological processes. Essentially, it selects core genes from a set of differentially expressed genes based on their similarity and visualises them using cloud diagrams. The analysis begins by converting the input gene list into Entrez gene IDs, which are then analysed using the GOSemSim package (version 2.22.0). The resulting output is sorted in descending order based on the average similarity of each gene to others, with the top genes being those that exhibit the highest similarity and are deemed key genes.

Statistical Methodology

Employing R software (version 4.2.1), we applied an array of MR techniques to dissect the causal interplay between 731 immunophenotypes and AD. IV heterogeneity was gauged using Cochran’s Q statistic. Upon rejection of the null hypothesis, we adopted a random effects IVW model. We scrutinised horizontal pleiotropy using the MR-Egger method, while MR-PRESSO addressed and rectified outlier-induced distortions. The <1> was used to account for multiple testing, with an FDR value <0.10 and p<0.05 set as the thresholds for significance. The research design and instrumental variable selection process are illustrated in Figure 1.

Results

Alignment of Genetic Variants with Causal Effect Estimates

Scatter plots demonstrate that all genetic variants align with the overall causal effect estimate and cluster closely around the regression line (Supplementary Figure 1). The funnel plot shows a symmetrical distribution, with horizontal lines for each SNP on both sides of the vertical line representing the overall causal effect estimate (Supplementary Figure 2). The LOOCV bar chart displays a unimodal distribution, with the peak value near the overall p-value (Supplementary Figure 3). These visual representations support the assumptions of the MR analysis, indicating that the SNPs used in this study are valid instrumental variables and confirming the robustness and reliability of our findings.

Causal Effect of Immunophenotypes on Measurement Outcomes

We conducted a two-sample MR analysis using the IVW method to assess the causal impact of immunophenotypes on AD events. Immunophenotypes with FDR below 0.10 were deemed to have significant causal associations. After applying multiple testing adjustments using the FDR approach, we identified four immunophenotypes that exhibit positive causal effects and one with negative causal effects on AD (Supplementary Table 1). These findings suggest both pathogenic and protective roles in AD. The pathogenic traits include: the median fluorescence intensity (MFI) of CD19 on transitional B cells, which acts as a marker for identifying the activation and differentiation status of these cells; the MFI of CD19 on immunoglobulin D (IgD)^- CD38dim B cells, providing insight into B cell subsets involved in immune responses; the MFI of CD3 on CD39⁺ CD4⁺ Treg cells, emphasising the role of Treg cells in modulating immune reactions and potentially impacting the integrity of the aortic wall; the MFI of CD3 on CD39⁺ activated Treg cells, serving as a marker for evaluating the activation and regulatory function of these cells. The sole protective trait is the absolute count of CD86⁺ myeloid dendritic cells, which underscores the importance of antigen-presenting cells in the pathogenesis of AD. Figure 2 presents a forest plot detailing the ORs, 95% CI, and FDR values for these immunocellular traits, as calculated by the IVW method.

Sensitivity Analysis

For each immune cell trait, we employed multiple MR methods to validate the causal associations between the selected immune cell types and AD outcomes. The results consistently demonstrated statistically significant associations, with most p-values being below 0.05 across different MR methods, including IVW, IVW-radial, MR-Egger, weighted median and weighted mode approaches. Notably, a few methods yielded p-values around 0.1, indicating marginal significance (Figure 3 and Supplementary Table 2). Specifically, we found that the MFI of CD19 on IgD^- CD38dim B cells, the MFI of CD19 on transitional B cells, and the MFI of CD3 on CD39⁺ CD4⁺ Treg cells consistently showed p-values <0.05 across all six MR methods. In contrast, the MFI of CD3 on CD39⁺ activated Treg cells and the absolute count of CD86⁺ myeloid dendritic cells (MDCs) exhibited p-values around 0.1 in several MR methods, indicating marginal significance. Furthermore, the heterogeneity tests for these immune cell traits across all MR analyses yielded p-values >0.05, suggesting no significant heterogeneity (Supplementary Table 2). This consistency across different MR approaches reinforces the robustness of our findings and suggests that the observed associations are not driven by methodological differences. Additionally, we used the intercept from MR-Egger (Supplementary Table 2) and the global test from MR-PRESSO (Supplementary Table 3) to rule out the possibility of horizontal pleiotropy in these associations (p>0.05).

To determine whether these five immune cell traits exhibited horizontal pleiotropy, we conducted multivariable MR analyses. The results indicated that only the association for the absolute count of CD86⁺ MDCs was statistically significant, with a p-value less than 0.05. The other immune cell traits demonstrated p-values significantly less than 0.1, suggesting no significant causal associations (Figure 4). The reverse MR analysis did not identify any evidence of reverse causation, indicating that AD does not causally influence the selected immune cell traits (Supplementary Table 4).

Overall, these findings reinforce the robustness of our initial results, suggesting that the identified immune cell types may play a significant role in the pathogenesis of AD.

Pathway Enrichment Analysis

We further analysed the biological functions of eQTL genes that are causally linked to immune cell characteristics, with details available on request. Using heatmaps, we illustrated the tissue expression differences of these genes (Supplementary Figure 4) and described the results of tissue enrichment analysis (Supplementary Figure 5). Our findings underscore the heterogeneity in gene expression among different tissues and highlight significant arterial enrichments. This sheds light on possible mechanistic pathways that these immune cells may engage in within the context of AD.

Pathway analysis using the Metascape tool, as shown in Figure 5A, revealed that the positive regulation of immune response is the most significant biological process involving these causally immune cell trait-associated eQTL genes. Subsequently, we used the clusterProfiler package in R software to elucidate the specific molecular mechanisms of the eQTL genes. As shown in Figure 5B, we found that the primary biological processes involved in mononuclear cell and lymphocyte differentiations, as well as leukocyte activation involved in immune response, are closely associated with the development and progression of various cardiovascular diseases. Friends analysis identified signalling lymphocytic activation molecule family member 6 (SLAMF6) and CD28 as the two key genes linked to these pathways, due to their high similarity with other genes in the network (Figure 6).

Discussion

Causal Associations and Biological Significance

Our two-sample MR analysis using the IVW method found significant causal links between specific immunophenotypes and AD events. After adjusting for multiple tests using the FDR approach, we identified four immunophenotypes with positive causal effects and one with a negative effect on AD. These findings indicate both harmful and protective roles for these immunophenotypes in AD. The harmful traits include: MFI of CD19 on transitional B cells; MFI of CD19 on IgD-CD38dim B cells; MFI of CD3 on CD39⁺ CD4⁺ Treg cells; and MFI of CD3 on CD39⁺ activated Treg cells. The protective trait is the absolute count of CD86⁺ myeloid dendritic cells. Sensitivity analyses using various MR methods consistently showed significant associations for these immune cell traits, with most p-values below 0.05. Heterogeneity tests and pleiotropy assessments showed no significant issues. Multivariable MR analyses revealed that only the absolute count of CD86⁺ myeloid dendritic cells remained significant, suggesting it may be the most influential immunophenotype. Other traits did not show significant causal associations in the multivariable context. Reverse MR analysis confirmed no reverse causation. These results suggest that while initial univariate MR analyses indicated potential causal relationships for multiple immunophenotypes, multivariable MR analyses highlight the absolute count of CD86⁺ MDCs as the most robustly associated trait with AD. Further research is needed to validate these findings and explore other factors influencing these associations for a better understanding of immune cell traits and their impact on AD.

To further elucidate the biological underpinnings of these associations, it is essential to delve into the individual roles and mechanisms of each identified immunophenotype. The following sections will provide a detailed examination of the biological functions of the five key immune cell traits — transitional B cells, IgD- CD38dim B cells, CD4⁺ Treg cells, activated Treg cells, and MDCs — and their potential causal links to AD.

Transitional B Cells and IgD^--CD38dim B Cells

The MFI of CD19 on transitional B cells and IgD-CD38dim B cells emphasises their activation and differentiation status, showcasing their crucial roles in the immune response. Transitional B cells, transitioning from immature to fully mature B cells, play a vital role in early immune responses.²⁴ Their activation status, indicated by the CD19 marker,^25,26 suggests that disruptions in B cell maturation and function could contribute to the inflammatory processes that weaken the aortic wall, leading to AD. Similarly, IgD-CD38dim B cells, representing a more differentiated state involved in adaptive immunity, may serve as a critical checkpoint where immune dysregulation can occur.^27,28 The positive association of these B cell subsets with AD suggests that dysregulated B cell activation and differentiation may play a significant role in AD pathogenesis. Therapeutic strategies aimed at modulating these specific stages of B cell development could potentially stabilise the immune environment and prevent the progression of AD.

T Regulatory Cells

The MFI of CD3 on CD39⁺ CD4⁺ Tregs and CD39⁺ activated Tregs underscores their regulatory function and activation status, highlighting their pivotal role in maintaining immune homeostasis and preventing excessive inflammation.^29,30 Tregs, particularly those expressing CD39, are central to immune tolerance and regulation.^29,31,32 CD39 is an ectoenzyme that hydrolyses extracellular adenosine triphosphate, a pro-inflammatory molecule, into AMP, thereby exerting anti-inflammatory effects.²⁹ The positive causal effect of these Treg subsets on AD suggests that their dysregulation may lead to inadequate suppression of inflammatory responses, compromising the integrity of the aortic wall. Diminishing the function or numbers of CD39⁺ Treg cells could be a promising strategy to mitigate inflammation and preserve aortic integrity in AD patients. Additionally, understanding the genetic factors that influence CD39 expression and function could provide insights into individual susceptibility to AD and guide personalised therapeutic approaches.

CD86⁺ Myeloid Dendritic Cells

CD86⁺ MDCs are potent antigen-presenting cells that play a critical role in initiating and regulating immune responses.³³ The expression of CD86 on these cells indicates their activation status and ability to provide co-stimulatory signals necessary for T cell activation and survival.^33,34 The negative causal effect of the absolute count of CD86⁺ MDCs on AD suggests that these cells contribute to maintaining immune homeostasis and preventing excessive inflammatory responses that could damage the aortic wall. By presenting antigens and modulating T cell responses, CD86⁺ MDCs help regulate the immune environment, protecting against the pathological processes leading to AD. Enhancing the antigen-presenting capabilities of MDCs or boosting their numbers could fortify the immune system’s ability to detect and respond to early signs of aortic damage, thereby preventing the onset or worsening of AD.

From a translational clinical perspective, it is encouraging that ongoing clinical trials and research are exploring the role of immune cell types in the management of AD. Specifically, studies are examining the causal relationship between thoracic aortic aneurysm and immune cells, which is closely related to AD. These studies employ advanced genetic techniques to elucidate how specific immune cell phenotypes contribute to the disease.³⁵ Furthermore, research on cell-based therapies for aortic aneurysms, including the modulation of immune responses through stem/progenitor cells, is making progress.³⁶ These efforts aim to control inflammation and promote tissue repair, which are critical for AD management.

eQTL Genes and Molecular Pathways

Pathway enrichment analysis revealed that the eQTL genes associated with these immune cell traits are involved in key biological processes such as mononuclear cell and lymphocyte differentiation, and leukocyte activation. Notably, SLAMF6 and CD28 emerged as core genes within these pathways, indicating their pivotal roles in immune response regulation.

SLAMF6 is a receptor expressed on various immune cells, including T cells and natural killer cells. It plays a crucial role in modulating immune cell activation and proliferation.³⁷ The engagement of SLAMF6 with its ligands can enhance T cell receptor signalling, leading to heightened T cell responses.³⁸ This enhanced activation can contribute to the inflammatory milieu, which is a known risk factor for AD. SLAMF6 is particularly important for distinguishing between progenitor-exhausted and terminally exhausted CD8⁺ T cells, maintaining a pool of functional T cells during chronic infections and cancer.^39,40 Dysregulation in SLAMF6 c can result in hyperactive immune responses, contributing to vascular inflammation and remodelling seen in AD.⁴¹

CD28 is a well-known co-stimulatory molecule necessary for full T cell activation.⁴² It binds to its ligands CD80 and CD86 on antigen-presenting cells, providing the necessary second signal for T cell activation.⁴³ Without this co-stimulatory signal, T cells may become anergic or undergo apoptosis. CD28 signalling enhances the production of interleukin-2, a critical growth factor for T cells, and promotes their survival by upregulating anti-apoptotic proteins. This pathway is vital for the effective initiation and maintenance of adaptive immune responses. However, excessive or dysregulated CD28 signalling can lead to chronic inflammation and tissue damage, exacerbating the inflammatory processes that weaken the aortic wall and promote AD progression.⁴¹

Further exploring the downstream signalling pathways of SLAMF6 and CD28 could provide more insights. For example, SLAMF6 engagement leads to the activation of the SLAM-associated protein (SAP) and Fyn kinase pathways, integral for cytotoxic T cell and NK cell functions.^39,44 Similarly, CD28 signalling activates the PI3K-Akt and NF-κB pathways, crucial for T cell survival, growth and cytokine production.^44,45 Chronic activation of these pathways can sustain an inflammatory environment, promoting the recruitment and activation of additional immune cells to the aortic tissue, exacerbating tissue destruction and aneurysm formation.^46,47 Understanding these mechanisms could guide the development of targeted therapies to modulate immune responses and prevent AD. Moreover, advanced single-cell RNA sequencing (scRNA-seq) has enabled the detailed mapping of immune cell populations within affected aortic tissues. Such analyses have pinpointed subsets of T cells and other immune cell types that exhibit elevated expressions of SLAMF6 and CD28, which are linked to areas of pronounced inflammation and tissue damage.

Leveraging Immune-related Features and eQTL Genes for the Prevention of Aortic Dissection

Using the discovered immune-related traits and associated eQTL genes for screening high-risk populations or preclinical interventions for AD presents a promising approach. Genetic screening could be a pivotal strategy, where eQTL-based genetic tests are developed to identify specific eQTLs associated with immune cell traits linked to AD. Individuals carrying these genetic markers could be flagged as high risk and subjected to closer monitoring. Incorporating genetic screening into routine check-ups, especially for those with a family history of AD, along with genetic counselling, can help assess risk and guide preventive measures.

Biomarker identification through immune cell profiling using advanced techniques, such as flow cytometry can also play a crucial role. Elevated levels of specific immune cell traits, such as the MFI of CD19 on transitional B cells or CD3 on CD39⁺ Treg cells, could indicate a higher risk of AD. Regular monitoring of these biomarkers in high-risk individuals could detect early signs of immune dysregulation that may precede AD. Additionally, machine learning and AI can be leveraged to develop predictive models that integrate genetic, immunological, and clinical data to predict AD risk, thereby prioritising individuals for further testing and preventive measures.^3,48

Preclinical interventions may involve immunomodulatory therapies for high-risk individuals. Preventive immunotherapy that targets specific immune cell traits, such as modulating B cell activation or enhancing Treg cell function, could be considered. One promising strategy is to target B cell activation. Because CD19 plays a pathogenic role on transitional B cells and IgD-CD38dim B cells, developing CD19 inhibitors could reduce B cell activation and differentiation, potentially lowering AD risk. Additionally, B cell depletion therapies, such as rituximab, which modulate the immune response by depleting B cells, could be explored for their effectiveness in preventing AD. Enhancing Treg function is another viable strategy.⁴⁹ Expanding Treg cells with low-dose interleukin-2 therapy could improve their regulatory functions and maintain aortic wall integrity.⁵⁰ Furthermore, since CD39⁺ Treg cells are involved in AD, modulating CD39 activity to boost Treg function could be promising.³⁵ Another strategy is to enhance antigen-presenting cell activity. Stimulating CD86⁺ myeloid dendritic cells, which have a protective role, with agonists could improve antigen presentation and immune regulation.⁵¹ Future research and clinical trials will be essential to validate these approaches and assess their efficacy and safety in patients.

Clinical implementation of these strategies would involve establishing early detection programmes with screening protocols in clinical settings to identify high-risk individuals based on genetic and immunological markers. Early detection can lead to timely interventions and better outcomes. Patient education about the importance of regular screening and monitoring, especially for those with a family history of AD or known genetic predispositions, is crucial. Collaborative research through multicentre studies can validate the effectiveness of these screening and intervention strategies, accelerating the development of effective preventive measures. Designing and implementing clinical trials to test the safety and efficacy of preclinical interventions in high-risk populations will provide valuable data to guide clinical practice.

Challenges in Implementing Immune-Based Therapies for Aortic Dissection

Implementing the potential therapies based on our findings could face several significant challenges. First, the safety and efficacy of these therapies must be thoroughly evaluated. Therapies targeting immune cells, such as B cell depletion or Treg cell expansion, could lead to unintended immune suppression or activation, increasing the risk of infections or autoimmune diseases. Additionally, the effectiveness of these therapies may vary across different genetic backgrounds and populations, necessitating extensive clinical trials to ensure broad applicability.

Strengths and Limitations

A major strength of our study is the robust methodology employed, using multiple MR methods such as IVW, MR-Egger and weighted median approaches. This multi-faceted approach minimises biases and confirms the robustness of our causal associations. Additionally, the comprehensive pathway enrichment analysis integrates genetic, immunological and bioinformatics approaches, providing a holistic understanding of the molecular mechanisms underlying AD.

However, our study has certain limitations. A significant constraint is the inherent dynamic nature of immune cell populations, which are influenced by transient factors such as infections, stress or circadian rhythms. Despite the rigorous quality control measures applied to the original dataset, including processing blood samples within 2 hours of collection and using standardised protocols to minimise time-dependent artefacts, these methods cannot eliminate the impact of temporal variability. As our analysis relied on cross-sectional data, it may not fully capture the long-term steady-state immune profiles. Future studies using longitudinal data or repeated measurements are needed to better address these dynamic changes and validate the robustness of our findings. Second, our findings are based on genetic data from specific populations, which may limit their generalisability to other ethnic groups. Although methods were employed to detect and account for horizontal pleiotropy, residual pleiotropy cannot be entirely ruled out. Third, we were unable to further refine the phenotype classification for the selected GWAS dataset of AD due to data limitations. This broad definition encompasses various subtypes, which may include distinct genetic, congenital, or environmental aetiologies. We acknowledge that this limitation could introduce heterogeneity into our analysis and potentially affect the interpretation of causal relationships. Future studies using more granular phenotype data and advanced stratification methods are warranted to address this issue.

Furthermore, the study is primarily based on statistical associations, necessitating experimental validation of the identified pathways and genes to confirm their biological relevance. To achieve a balance between discovery and stringency, we selected a significance threshold of FDR<0.10 for mapping risk loci genes associated with causative immune cells. An FDR threshold can range from 0.05 (more stringent) to 0.20 (more lenient), and we selected FDR<0.10 for our study to include a broader set of potential risk genes for further pathway and mechanism exploration. This exploratory approach allows us to gain a more comprehensive understanding, although it may result in some false positives. Future studies should aim to replicate our findings in diverse populations and explore the clinical applicability of these immune cell traits and the mapped key genes in AD management. This will help ensure that our results are broadly applicable and can inform effective therapeutic strategies.

Conclusion

Our MR analysis highlights significant causal relationships between specific immune cell traits and AD. The identification of both pathogenic and protective immunophenotypes underscores their potential clinical implications for risk stratification and targeted therapeutic interventions in AD management. The validated associations and enriched biological pathways, including key genes such as SLAMF6 and CD28, present promising avenues for future research. These findings provide a foundation for translating genetic insights into practical strategies for improved patient care, though further investigations are warranted to corroborate these results.

Click here to view Supplementary Material.