To the best of our knowledge, this is the first study with sufficient statistical power to investigate the associations of genetically determined ICAM-4 levels with the risk of ischemic stroke and its subtypes. In this MR study with 446,696 European participants, we found that genetically determined high ICAM-4 levels were associated with increased risks of ischemic stroke and CES, but not LAS or SVS. Sensitivity analyses using a series of MR methods further confirmed these identified associations.
As the key molecules involved in progression of the ischemia, the upregulation of ICAMs was able to enhance leukocyte-endothelial cell interactions and induce neutrophils infiltrate into damaged brain tissue, thereby aggravating the damage of blood-brain barrier.10 ICAM-4, an important member of the ICAMs, was well-established to be implicated in hemostasis and thrombosis.6,7 Data from animal studies had revealed that ICAM-4 contributed to the adhesion of endothelial cells, thrombosis, and vaso-occlusion through interacting with activated platelets and leukocytes.31,32 Vivian et al. found that ICAM-4 could induce massive erythrocytes incorporation into thrombus and activate platelets, while blockading ICAM-4 could result in a reduction of fibrin and thrombus.8 Therefore, it has been suggested that ICAM-4 may be associated with the development of thrombosis and ischemic stroke, but the population-based evidence is limited so far.
We conducted a systematical MR study to investigate the associations of genetically determined ICAM-4 levels and the risks of ischemic stroke and its subtypes. In this MR study, genetically determined high ICAM-4 levels were observed to be associated with an increased risk of ischemic stroke, suggesting that it might be a promising predictive marker for ischemic stroke. Besides, in the analysis of ischemic stroke subtypes, we found that there were possible mechanism-specific detrimental effects of genetically determined ICAM-4 on CES but not LAS or SVS. Given that ischemic stroke subtypes may differ for the genetic pathophysiological mechanisms,33 CES had a higher inflammatory environment, macrophage, and platelet content in thrombus than the other stroke subtypes.34,35 Therefore, we speculated that elevated ICAM-4 levels were significantly associated with increased risks of ischemic stroke and CES via mediating aggregation and abnormal adhesion of inflammatory cells, macrophages, and platelets. Further studies are warranted to explore the detailed mechanism underlying the association of ICAM-4 with ischemic stroke and CES.
Our findings have significant public health and clinical implications. The present MR study was the first to provide population-based evidence for the associations of ICAM-4 levels with the risk of ischemic stroke and its subtypes from a genetic point of view. Collectively, our study showed that elevated ICAM-4 levels increased the risk of ischemic stroke and CES, suggesting that ICAM-4 might be acted as a promising biomarker to identify high-risk individuals for active monitoring and early intervention of ischemic stroke. In addition, it is of clinical interest to explore whether targeting ICAM-4 or its downstream effectors could reduce the risk of ischemic stroke, especially CES.
The present study had several methodological strengths. First, MR design follows the genetic rule that parental alleles are randomly assigned to offspring and possesses reasonable causal order.11,12 Therefore, the implementation of MR approach in this study diminished the interference of confounding factors and reverse causation on the results, which might be more convincing than observational studies.11,12 In addition, we used the most comprehensive and the largest available GWASs about ICAM-4 levels, ischemic stroke, and its subtypes,18,19 which enabled us to provide a valid appraisal of the association with high the statistical power. Finally, the significant associations observed in this MR study were subjected to multiple corrected and a series of sensitivity analyses further confirmed our findings.
Our study has several limitations that need to be interpreted. Firstly, MR analysis might be influenced by instrument bias and potential pleiotropy. However, the F-statistic for the genetic instruments in the present study was greater than 10, suggesting that there was no weak instrument bias. Furthermore, the MR-Egger regression suggested no directional pleiotropy for these associations. Therefore, there was no pleiotropy in this MR study. Secondly, this MR study estimated the lifetime effect of plasma ICAM-4 in the risk of ischemic stroke and ischemic stroke subtypes, so the results should not be directly extrapolated to assess the effect of any potential clinical intervention targeting ICAM-4. Finally, the summary GWAS data we used merely concerned European individuals, so we should cautiously utilize our conclusion in racially and ethnically diverse populations. However, this restriction decreased the possibility of spurious associations due to population stratification bias, and further studies are needed to confirm our findings among individuals of non-European ancestry.