Immunoglobulin E Sensitization to Mammalian Oligosaccharide Galactose-α-1,3 (α-Gal) Is Associated With Noncalcified Plaque, Obstructive Coronary Artery Disease, and ST-Segment–Elevated Myocardial Infarction

The identification and treatment of modifiable risk factors for coronary artery disease (CAD) has substantially improved outcomes for CAD in the community.¹ Despite this, at the individual level patients presenting with life threatening atherosclerosis driven events without adequate explanation are not uncommon. This is highlighted by the not insubstantial proportion of patients with myocardial infarction (MI) who develop unexpected life-threatening CAD despite the absence of any of the Standard Modifiable Cardiovascular Risk Factors (hyperlipidemia, hypertension, diabetes, and smoking). Twenty percent of ST-segment–elevated myocardial infarction (STEMI) patients included in a large international meta-analysis (74 147) had zero Standard Modifiable Cardiovascular Risk Factors (SmuRFs),² with data from 2 Australian cohorts demonstrating a significant increase in the proportion of without any standard modifiable cardiovascular risk factors (SmuRF-less) MI patients over recent decades.^3,4 Additionally, a recent retrospective analysis of the SWEDEHEART registry (Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies; 62 048) demonstrated that SmuRF-less STEMI patients have a ≈50% higher early mortality rate compared with STEMI patients with ≥1 SMuRF.⁵ This archetypal patient group highlights the importance of continued efforts to discover new mechanisms contributing to individual susceptibility to CAD, as well as considering potential environmental contributors.

The integral role of inflammation in the pathogenesis of atherosclerosis is well established, with causality confirmed by the CANTOS study’s (Canakinumab Antiinflammatory Thrombosis Outcome Study) demonstration that blockade of IL-1β (interleukin 1β) reduces cardiovascular events in individuals with a previous MI and elevated C-reactive protein.⁶ Elevated serum immunoglobulin E (IgE), classically involved in allergic inflammation, and early-phase responses, has been associated with atherosclerosis.^7–9 IgE-sensitization to the mammalian oligosaccharide galactose-α-1,3-galactose (α-Gal) can occur after an individual is bitten by certain species of tick such as Ixodes holocyclus.¹⁰ α-Gal sensitization has been detected across all continents in tick-endemic regions.¹¹ Oligosaccharide galactose-α-1,3-galactose (α-Gal) is ubiquitous in mammals other than humans and present in other species including parasites. It is plausible that human IgE α-Gal sensitization may occur in response to exposures other than tick bite, however, such sensitizations have not been well described to date. A small proportion of sensitized individuals develop an anaphylactic response to mammalian meat or cetuxumab (a chimeric mouse–human IgG1 monoclonal antibody) infusion, however, there is no data to date to suggest that such exposures in nonsensitized individuals can result in IgE α-Gal sensitization.^11–13 Subacute and chronic sensitization occurs more frequently, activating systemic immune pathways.^11,14 The potential role for this in atherosclerosis was suggested by recent data showing an association between α-Gal sensitization and CAD, particularly unstable plaque.¹⁵

CT coronary angiography (CTCA) provides a powerful opportunity to examine novel mechanisms and markers of CAD, providing a noninvasive means to quantify and characterize atherosclerotic disease. BioHEART-CT is a large cohort study with CTCA imaging and stored blood samples. Using this resource, we examined 3 hypotheses: that α-Gal sensitization is associated with the presence of CAD, in particularly noncalcified (soft) plaque; that α-Gal sensitization is associated with greater plaque burden; and that α-Gal sensitization is associated with acute STEMI.

Methods

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

The BioHEART-CT Cohort

BioHEART-CT is a multicentre, prospective cohort study of patients with or at risk of CAD (ACTRN12618001322224).¹⁶ The study complies with the Declaration of Helsinki and has been approved by the Northern Sydney Local Health District Human Research Ethics Committee (HREC/17/HAWKE/343). As described in detail previously,¹⁶ the inclusion criteria are patients referred for CTCA for suspected CAD; >18 years old; able to provide informed consent. Exclusion criteria are patients unable to provide informed consent; and patients unwilling or unable to participate in ongoing follow-up. Patients with a prior history of coronary stents or bypass grafts were excluded from this analysis. The first 1056 participants meeting the inclusion/exclusion criteria were included in this study. The participants were recruited between November 2015 and June 2019 in Sydney, Australia. Baseline clinical data were obtained by questionnaire and review of medical records.

CTCA Acquisition and Analysis

Imaging was acquired on 256 slice scanners using standardized clinical protocols. Radiation dose was minimised in line with current recommendations.^17,18 CTCAs were analyzed using the 17-segment model.¹⁹ Each segment was scored according to degree of stenosis (0%, ≤25%, 26%–50%, 51%–75%, 76%–90%, 91%–99%, 100%) and composition of plaque (calcified/soft/mixed). Segmental scores were aggregated using validated segmental weightings to obtain a Gensini score representing the total amount of plaque present.²⁰ The presence of obstructive CAD was recorded if any vessel had a stenosis >50% (O50). CTCA scoring was performed by clinicians trained by Royal Australian and New Zealand College of Radiologists accredited CTCA specialists. Each score was independently reviewed by a second clinician in conjunction with the formal report. Interoperator disagreements were adjudicated by a Royal Australian and New Zealand College of Radiologists specialist with Level B accreditation (highest level). Clinicians performing the CTCA scoring were blinded to the α-Gal assay results.

A soft (noncalcified) plaque score (SPS) was derived from the weighted aggregate of the segmental soft plaque value using the plaque composition multiplier (×1 for calcified only, ×2 for mixed plaque, ×3 for soft plaque only).

Coronary artery calcium scores (CACS) were generated using the validated Agatston method.²¹

The BioHEART-MI Cohort

BioHEART-MI recruits patients admitted with suspected acute MI at the time of index coronary angiography. Inclusion criteria are patients undergoing emergent coronary angiography with confirmed culprit plaque; age 18 years or older; and willing and able to provide informed consent. Exclusion criteria are the same as outline above for BioHEART-CT. For this analysis, we enriched for SMuRF-less STEMI patients, taking the first 48 SMuRF-less patients along with 52 STEMI patients with ≥1 SMuRF. BioHEART-MI participants included in this study were recruited from the same center as the BioHEART-CT participants included in this study and resided in the same geographic region of Sydney, Australia.

Sample Processing and IgE Assay

Specific IgE antibodies to α-Gal were measured in thawed serum using the commercially available automated assay using nGal-alpha-1,3-Gal (alpha-Gal) from bovine thyroglobulin (ImmunoCAP 1000 [TM] Thermofisher Phadia). The assay was calibrated using standard concentrations, 0.001 kU/L, 0.35 kU/L, 0.70 kU/L, 3.50 kU/L, 17.5 kU/L, 100 kU/L before each run. One hundred samples were assessed in each batch including a positive and negative control. A cutoff of ≥0.10 kU/L was used to define α-Gal sensitization.²² Those performing the IgE α-Gal assay were blinded to the CTCA scores.

Standard Modifiable Cardiovascular Risk Factors

A patient was considered to have a known history of hypertension, hypercholesterolemia, or diabetes if they identified as having a history of the same or if they were on pharmacological agents for the condition. A significant smoking history was defined as ≥10 packet year history of smoking or current smoking. Additionally, where available, a total cholesterol >5.5 mmol/L or LDL (low-density lipoprotein) >3.5 mmol/L was included in the definition of hypercholesterolemia, an HbA1c (hemoglobin A1c) >6.5% was included in the definition of diabetes. Significant family history of premature CAD was defined as ischemic heart disease in a first degree family member younger than 60 years old.

Statistical Analysis

Pearson χ² test was used to compare categorical variables between groups. Student t tests were used to compare continuous variables between groups with normal distribution and equal variance as assessed by the Shapiro-Wilk test and Levene test of equal variance respectively. Non-normally distributed continuous variables were assessed using Mann-Whitney U test. CAD severity including Gensini and soft plaque scores are not normally distributed, and due to the large proportion of zero values are not amenable to normalizing transformations.^23–26 However, after exclusion of cases with no disease, the distribution of log10 transformed CAD scores were approximately normal.

We performed univariate binary logistic regressions between the binary dependent variables Gensini score >0 and SPS >0, 050 and the independent variables. Univariate linear regression models were performed to assess the association between the dependent variables log₁₀Gensini and log₁₀SPS and independent variables. The independent variables tested were α-Gal sensitization, age, sex, body mass index, hypertension, significant smoking history, diabetes, hypercholesterolemia, index of relative socioeconomic disadvantage decile, and ethnicity. Covariates with P<0.10 on univariate testing were included in multivariable models, using a backward stepwise selection method. A post hoc analysis assessing for association of the presence of CAD with IgE α-Gal in women and men separately was performed and is presented in the Supplemental Material. Propensity score matching using logistic regression was performed to obtain an age and sex matched stable CAD group and an age and sex matched healthy control group from the BioHEART-CT cohort (confirmed no CAD) for comparison with the BioHEART-MI STEMI group. All 100 STEMI patients were able to be age/sex matched with stable CAD patients; however, only 78 of the STEMI group were able to be matched with healthy no CAD controls. Regressions were performed within the framework a generalized estimating equation. Analyses were performed using IBM SPSS statistics, version 26.0 (IBM Corp, Armonk, NY).

Patient and Public Involvement

We conducted a patient focus group to assess willingness of patients to participate in the trial and to identify potential barriers to participation, the results of which were used to inform study design. We invite study participants to events aimed at the general public to update them regarding progress and findings of the BioHEART study. We are also developing a study website and newsletter to broaden the reach of the study and to enhance communication with participants.

Results

Presence of Disease (Hypothesis 1)

The clinical and demographic characteristics are presented in Table 1. Eleven percent of the cohort were sensitized to α-Gal. A small difference in age between the α-Gal sensitized group and nonsensitized group was present (64 [interquartile range, 57–72] versus 61 [interquartile range, 52–69], P=0.02). There were no other differences in demographic or baseline characteristics (Table 1). α-Gal sensitized individuals had a 20% higher rate of CAD compared with nonsensitized individuals (hazard ratio, 1.2 [95% CI, 1.1–1.4], P<0.001). This difference was more marked for noncalcified (soft) plaque, where α-Gal sensitized individuals had 1.3-fold higher incidence of soft plaque (68% versus 52%, P=0.002, Figure 1).

• Download figure
• Download PowerPoint

The association of α-Gal sensitization with CAD was confirmed in the univariate logistic regression model (odds ratio [OR], 1.99 [95% CI, 1.25–3.16], P=0.004), however, did not remain significant after adjustment for cardiovascular risk factors in the prespecified multivariable model (Figure 2). In contrast, α-Gal sensitization was associated with the presence of noncalcified plaque in the univariate logistic regression model (OR, 1.92 [95% CI, 1.27–2.92], P=0.002) and this relationship remained after adjustment for risk factors in the multivariable model (OR, 1.62 [95% CI, 1.04–2.53], P=0.03).

• Download figure
• Download PowerPoint

Severity of Disease (Hypothesis 2)

We evaluated the relationship of α-Gal sensitization to disease severity. The average Gensini score (reflecting total plaque burden) was 86% higher in the α-Gal sensitized group, and the average soft plaque score was 300% higher than nonsensitized individuals (P<0.0001 for both; Figure 1). The incidence of obstructive CAD was 2.7-fold higher in the α-Gal sensitized patients compared with nonsensitized patients (49% versus 18%, P<0.001; Figure 1). α-Gal sensitization was significantly associated with obstructive CAD in both univariate and multivariable models (OR, 2.33 [95% CI, 1.52–3.57], P<0.001; and OR, 2.05 [95% CI, 1.29–3.25], P=0.002, respectively; Figure 2). The association of α-Gal sensitization and Gensini score was confirmed in the univariate linear regression model but did not persist after adjustment for other risk factors in the multivariable model. In contrast, the relationship between SPS and α-Gal sensitization was significant in both univariate and multivariable models (standardized β-coefficient, 0.10; P=0.02, Table 2).

We observed a strong association between logCACS and both the presence and amount of soft plaque (Pearson correlation coefficients 0.66 and 0.56, respectively, P<0.001 for both), as well as the presence of obstructive CAD (Pearson correlation coefficient, 0.55, P<0.001). To explore the independent association between α-Gal sensitization and the other metrics of CAD (SPS and O50), we repeated stepwise multivariable logistic and linear regression models for soft plaque and obstructive CAD incorporating logCACS as a covariate (Figure 2 and Table 2, multivariable models-2). The strength of the association (independent of traditional risk factors) between α-Gal sensitization and both of these disease measures was even stronger following adjustment for logCACS. The OR conferred by α-Gal sensitization status for the presence of soft plaque increased from 1.62 (95% CI, 1.04–2.53; P=0.03) to 2.08 (95% CI, 1.21–4.60; P=0.008) after adjustment for CACS. For obstructive CAD, the OR increased from 2.05 (95% CI, 1.29–3.25; P=0.002) to 2.35 (95% CI, 1.33–4.15; P=0.003; Figure 2). In addition to the stronger association, the performance of the models also substantially improved, shifting the multivariable logistic regression model R² from 0.23 to 0.54 for soft plaque, and 0.20 to 0.47 for obstructive CAD, and the multivariable linear regression model R² from 0.05 to 0.20 for soft plaque burden (Figure 2 and Table 2).

IgE α-Gal Sensitization Is Associated With Acute STEMI (Hypothesis 3)

To further examine the hypothesis that α-Gal sensitization is associated with plaque instability, we compared the rates of α-Gal sensitization in 100 STEMI patients from BioHEART-MI versus age and sex-propensity matched patients from the BioHEART-CT cohort. Baseline demographic and risk factor characteristics for the STEMI cohort and the age/sex matched groups are presented in Table 3. The rate of α-Gal sensitization was 2.2-fold higher in the patients suffering STEMI compared with the matched stable CAD patients from the same geographic area (20% versus 9%; P=0.03; Figure 3). This difference was dramatically higher (12.8-fold greater rate of α-Gal sensitization) when the STEMI group was compared with individuals from the same geographic area confirmed to have no CAD on CTCA (17 % versus 1.3%; P=0.01; Figure 3).

• Download figure
• Download PowerPoint

Discussion

This large study with over 1150 participants provides strong supportive evidence for a clinically important effect of α-Gal sensitization on the development of CAD and acute MI.¹⁵ We demonstrate that α-Gal sensitization is independently associated with soft plaque burden and the presence of obstructive disease. Furthermore, a dramatically higher prevalence of α-Gal sensitivity in STEMI patients compared with matched patients with no CAD or stable CAD highlights a likely involvement in the pathogenesis of plaque vulnerability and acute atherosclerotic events. These findings are both consistent with and build upon the previous smaller intravascular ultrasound based analysis that demonstrated an association between α-Gal sensitization and greater fibrofatty and necrotic plaque content.¹⁵ These findings have substantial public health implications given the high rates of α-Gal sensitization in tick-endemic areas globally.

How Representative Are Our Findings to Global Cardiovascular Health?

The BioHEART-CT and BioHEART-MI cohorts included in this study were recruited in Sydney, Australia, where there are a substantial number of suburbs where Ixodes holocyclus ticks responsible for α-Gal sensitization are endemic.¹¹ While these findings may be less relevant to nonendemic geographic regions, α-Gal sensitization secondary to a number of tick species have been identified in vast areas across every populated continent in the world.^11,14 The reported prevalence of α-Gal sensitization varies between geographic regions, thought mainly to be related to the relative tick endemicity of the region, for example α-Gal syndrome is relatively common in Southern Sweden where Ixodes ricinus is endemic, however, there are no reports of α-Gal sensitization in Northern Sweden where Ixodes ricinus is not found.^14,22,27 In tick endemic areas, α-Gal sensitization is present in as much as 35% of the population.²⁸ The 11% IgE α-Gal sensitization identified in our cohort is similar to the 10% identified using the same diagnostic threshold in asymptomatic blood donors in Stockholm, Sweden which is also a tick endemic region.²⁹

This study raises a number of important questions in regard to both clinical practice and public health policy. What can we do to minimize the risk of tick bites in the community, particularly in individuals with recreational or occupational exposure? A German study identified a 2.5-fold higher rate of α-Gal sensitization in forest service employees compared with a residential population.²⁸ Public policy and interventions that reduce exposure to tick bites should reduce α-Gal sensitization rates, with a potential beneficial impact on reducing CAD. Further community education may be required in tick endemic areas encouraging long sleeved shirts and pants with fitted cuffs to be worn in forest or grassed areas, preferably treated with. permethrin, with N,N-Diethyl-m-toluamide insect repellent applied at clothing skin interfaces.¹¹ There have been anecdotal reports of α-Gal syndromes in first degree relatives which may indicate a genetic predisposition to α-Gal sensitization; however, this may also be due to common environmental exposures and warrants further investigation.

Prospective studies are required in tick endemic areas to better understand the association and potential causal role of α-Gal sensitization in the risk of MI; however, the strong association with noncalcified plaque and the higher incidence of sensitization in STEMI patients provides modest evidence to support screening for IgE α-Gal in communities where ticks are endemic, or in individuals with high exposure risk through occupation or travel. However, it does raise further questions about what should be done differently if sensitization is detected.

Avoidance of recurrent tick bites does reduce IgE α-Gal levels and over time many mammalian meat allergy sufferers are able to return to eating meat; however, whether this has any impact on the risk to CAD is unknown.³⁰ A prospective cohort study of patients in endemic regions who are α-Gal sensitized may be beneficial to clarify the value of preventative measures in the context of atherosclerosis. Additionally, a prospective study of this type could examine the potential value of screening for subclinical atherosclerosis using CTCA in α-Gal sensitized individuals and determining the role for established primary prevention pharmacotherapy to help guide practice.

Inflammation, classically mediated by macrophages, neutrophils, and Th1 lymphocytes, is an important driver of both the development and progressions atherosclerosis.^6,31 There is mounting evidence for an important role of effector cells of allergic inflammation in the pathogenesis of atherosclerosis, which is likely relevant to the findings of our study linking α-Gal sensitization to active atherosclerosis and STEMI.³¹ Elevated circulating eosinophil counts secondary to allergy have been associated with MI, however, histological studies have not identified significant eosinophil recruitment to atherosclerotic plaques. Eosinophil derived mediators, in particular ECP (eosinophil cationic protein), on the contrary have been implicated in proatherogenic pathways, through the upregulation of intercellular adhesion molecule 1 expression on endothelial cells and promoting monocyte adhesion.³¹ Additionally the release of vasoactive inflammatory molecules, histamine, and leukotrienes, by activated mast cells is implicated in atherogenesis through increasing endothelial permeability, promoting the entry of circulating inflammatory cells and lipoproteins into the vessel wall.^31,32 Additionally, rare inherited hypereosinophyllic syndromes have been associated with increased thrombotic risk.³³

Our study has many strengths, including the size of the cohort and the detailed CAD phenotyping by CTCA. However, there were some limitations. It is not known what proportion of the cohort had experienced symptoms secondary to IgE α-Gal sensitization, and we were unable to determine whether the severity of clinical α-Gal allergy is associated with susceptibility to atherosclerosis. Based on other cohorts and clinical experience, it is likely that <10% of α-Gal sensitized individuals in the study had experienced symptoms consistent with an α-Gal allergy.²⁸ This is an observational study and while efforts were made to adjust for potential confounders, there may be additional unidentified confounders that were not controlled for. While the α-Gal sensitized group was slightly older than nonsensitized, the associations identified remained in the multivariable models adjusted for age. We were not able to assess for genetic/inherited factors that may contribute to the development of CAD other than self-reported significant family history of CVD. Sex specific analyses are presented in the Supplemental Material although noting that this study was under powered to detect sex specific associations. A potential confounder is the known association of total IgE with atherosclerosis, which we did not measure. However, Wilson et al¹⁵ have previously established the specificity of the relationship between IgE α-Gal and CAD. Specifically, they demonstrated that the relationship between IgE α-Gal and CAD is both independent of and stronger than the association between total IgE and CAD. Additionally, specific IgE antibodies to a panel of inhalants and peanuts were not associated with CAD. It has thus been suggested that the relationship between total IgE and CAD may be predominantly driven by IgE α-Gal specific antibodies. There is also no suggestion the IgE α-Gal is related to other food or aeroallergens. We did not collect data regarding prior tick bites, or information regarding risk of tick exposure other than location (postcode) of current residence.

Conclusions

We report that a specific IgE antibody response to α-Gal (induced following tick bites¹¹) is independently associated with the presence and amount of noncalcified atherosclerotic plaque and obstructive CAD in patients being investigated for stable CAD. Additionally, the proportion of acute STEMI patients who were sensitized to α-Gal in a tick endemic area is significantly higher than in an age and sex matched stable CAD and healthy control cohorts from the same geographic areas. These findings warrant further investigations to unravel potential disease mechanisms and pathways, as well as to elucidate potential roles in cardiovascular risk assessments and prevention programs, particularly in tick endemic areas.

Article Information

Acknowledgments

We would like to acknowledge the dedicated radiology department, cardiology ward, and cardiac catheter laboratory staff, together with the BioHEART recruitment teams at the recruitment sites for their support and invaluable contribution to the BioHEART study, the staff at Douglass Hanly Moir Pathology Macquarie Park Laboratory who performed the specific IgE testing, and Dr Richard Boyle, Primary Care Physician for his encouragement to undertake this study. The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy and ethics requirements. S.T. Vernon, M.R. Ward, D.S. Celermajer, S.J. Nicholls, S.A. van Nunen, S.M. Grieve, G.A. Figtree conceived and designed the study. S.T. Vernon, K.A. Kott, T. Hansen, R. Bhindi, P.S. Hansen, M.R. Ward, K.W. Baumgart, S.A. van Nunen obtained the data. S.T. Vernon and S.M. Grieve performed the statistical analyses. S.T. Vernon, K.A. Kott, S.A. van Nunen, S.M. Grieve, G.A. Figtree interpreted the data. All authors critically reviewed the manuscript and provided final approval for submission.

Supplemental Materials

Tables S1–S6

Footnotes