Interleukin‐18 in patients with acute coronary syndromes

Abstract Background We aimed to assess associations between circulating IL‐18 concentrations and cardiovascular outcomes in patients with acute coronary syndromes (ACS). Hypothesis and Methods Plasma IL‐18 concentrations were measured at admission, discharge, 1 month, and 6 months in patients with ACS in the PLATelet inhibition and patient Outcomes (PLATO) trial. Associations with outcomes were evaluated with Cox regression models on the composite of CV death, spontaneous myocardial infarction (sMI), or stroke; and on CV death or sMI separately, including adjustment for clinical risk factors and biomarkers (cTnT‐hs, NT‐proBNP, cystatin C, CRP‐hs, and GDF‐15). Results Median IL‐18 concentrations at baseline, discharge, 1 month, and 6 months were 237, 283, 305, and 320 ng/L (n = 16 636). Male sex, obesity, diabetes, and plasma levels of cystatin C, GDF‐15, and CRP‐hs were independently associated with higher IL‐18 levels. Higher baseline IL‐18 levels were associated with the composite endpoint and with CV death (hazard ratio [HR] 1.05, 95% confidence interval [95% CI] 1.02‐1.07 and HR 1.10, 95% CI 1.06‐1.14, respectively, per 25% increase of IL‐18 levels). Associations remained significant after adjustment for clinical variables but became non‐significant after adjustment for all biomarkers (HR 1.01, 95% CI 0.98‐1.04 and HR 1.04, 95% CI 1.00‐1.08, respectively). There were no associations with sMI. Conclusions In ACS patients, IL‐18 concentrations increased after the acute event and remained increased for 6 months. Baseline IL‐18 levels were significantly associated with CV mortality, independent of clinical characteristics and indicators of renal/cardiac dysfunction but this association was attenuated after adjustment for multiple biomarkers.


| INTRODUCTION
Inflammation is important for the development of atherosclerosis and inflammatory mediators synergistically amplify traditional risk factors of cardiovascular (CV) disease. 1 An increased inflammatory activity is associated with progression of stable coronary disease but also the future risk of acute coronary syndromes (ACS). [1][2][3] The role of inflammatory mediators in ACS is however less understood and routine measurement of inflammatory markers is not supported by international guidelines. 4 Interleukin-18 (IL-18), a member of the IL-1 cytokine superfamily, is a pleiotropic inflammatory cytokine expressed in several cells including macrophages and endothelial cells. 5 Increased circulating IL-18 concentrations have been associated with coronary artery disease and an increased risk for cardiovascular (CV) morbidity and mortality. 6,7 IL-18 has been detected in atherosclerotic plaques and has been proposed to destabilize coronary plaques leading to subsequent thrombotic events. 3 Nonetheless, the importance of circulating IL-18 in patients with ACS is not well documented. 8 The PLATelet inhibition and patient Outcomes (PLATO) trial included patients with ST-elevation and non-ST-elevation ACS, and showed that ticagrelor was superior to clopidogrel treatment in preventing CV mortality, myocardial infarction (MI), or stroke. 9,10 In the current biomarker substudy, we aimed to (a) evaluate the association between IL-18 concentrations and CV outcomes and bleeding and (b) explore the IL-18 concentrations during follow-up, and (c) assess possible interactions with randomized treatment (ticagrelor or clopidogrel).

| Design and study population
The global, randomized, placebo-controlled PLATO trial enrolled 18 624 patients with either ST-elevation ACS or non-ST-elevation ACS (Clinical Trial Registration: ClinicalTrials.gov NCT00391872). 9,10 The enrolled patients received optimal medical therapy including aspirin, and optional invasive strategy, and were randomized to either clopidogrel or ticagrelor treatment. 9,10 The patients were recruited between October 2006 and July 2008 and were subject to blood sampling as a part of a predefined biomarker substudy program. 9,10 The overall aims and details of the biomarker substudy program have previously been presented. 9 All participants provided written informed consent and the study complied with the declaration of Helsinki. The study was approved by all local Ethics Committees and Institutional Review Boards.

| Endpoint definition
The pre-specified primary composite endpoint of the present substudy was the combination of CV death, stroke or spontaneous MI (sMI) within 1 year of follow-up. 11 CV death and sMI were also separately evaluated, and secondary outcomes also included PLATO-defined non-CABG major bleeding. 9 Stroke was not separately evaluated, due to few events. All endpoints in the PLATO trial were centrally adjudicated by an independent and blinded clinical event adjudication committee in order to sub classify causes of death and to subdivide types of myocardial infarctions, stroke, and bleeding events. 9,11 CV death was defined as sudden death or death with no clear attributable non-cardiovascular cause. Spontaneous myocardial infarction was defined in accordance with the universal definition of myocardial infarction, hence a non-procedural, non-fatal, MI type 1. 11 The definition of stroke has previously been described. 9 PLATO major bleeding was defined as fatal or intracranial bleeding or requiring two units of blood transfusion or with a drop in hemoglobin of >5 g/dL. 9

| Sampling and laboratory analysis
Baseline venous blood samples were obtained at the time of randomization within 24 hours of admission, and prior to the administration of study medication. In addition, a predefined substudy program with serial blood sampling was conducted at selected sites in order to obtain samples from 4000 patients at discharge and after 1 month after randomization and from at least 3000 of these patients also at 6 months. Patients at these selected sites were continuously invited to participate in the substudy until it was estimated that at least 3000 patients would be available for blood sampling at 6 months. Patients with a baseline blood sample and at least one additional blood sample during follow-up were included in the serial biomarker analyses.
The venous blood was centrifuged and plasma samples were The growth differentiation factor-15 (GDF-15, ng/L) concentration was determined with Elecsys electrochemiluminescence immunoassay on a Cobas Immunoanalyzer system (Roche Diagnostics, Rotkreuz, Switzerland). 12 White blood cells (WBC) and high-sensitivity C-reactive protein (CRP-hs) were locally analyzed at site. Crude event rates at 1-year were estimated by IL-18 quartile groups. The association between IL-18 concentrations (ln transformed), on admission, with the composite primary endpoint of CV death, sMI or stroke, and secondary endpoints of CV death alone, sMI alone and non-CABG major bleeding were assessed by multivariable Cox proportional hazards models. Similarly, the associations between IL-18 quartile groups were examined, with the lowest quartile group as reference.   The variables significantly associated with increasing IL-18 concentration at baseline were male sex, age, diabetes, BMI > 30 kg/m 2 and increasing levels of the biomarkers NT-proBNP, cTnT-hs, CRP-hs, and cystatin C (Table 1).
Event rates for CV death alone per increasing quartile groups of IL-18 were 3.2%, 3.4%, 4.4%, and 5.3% (Table 3). Multivariable adjusted HR per 25% increase in IL-18 concentration on CV death was significantly associated with outcome, even after the adjustment for randomized treatment, clinical variables and the biomarkers cystatin C, cTnT-hs, and NT-proBNP. However, after additional adjustment with the biomarkers WBC, GDF-15 and CRP-hs in model 4, the association between IL-18 and CV death was insignificant ( Table 3, Figure 1).

| Interaction between IL-18 and the effects of randomized treatment on outcome
There was no significant interaction between the IL-18 quartile groups and the effects on outcomes by the randomized treatment.
The associations between IL-18 levels and the effects of randomized treatment (ticagrelor vs clopidogrel) were also investigated by restricted cubic splines ( Figure S1A, B, C and D). There were similar associations between IL-18 levels and outcomes irrespective of treatment with ticagrelor or clopidogrel, except for sMI where there was a trend for increasing events with increasing IL-18 level in the clopidogrel group but not the ticagrelor group.

| IL-18 concentrations and variables associated with IL-18 concentrations during follow-up
Patients with a baseline IL-18 sample and at least one subsequent follow-up sample (at discharge, 1 month, or 6 months) were included in the serial biomarker subset (n = 4583) ( Table 4). There was a significant increase in IL-18 concentrations during the first month, with a similar level at 1 and 6 months after the index event in ACS survivors ( Table 4).
The increase in IL-18 concentrations was observed irrespective of randomized treatment ( Table 4).
The following variables were significantly associated with the IL-18 concentration at three out of the four occasions during follow-up: male sex, obesity, diabetes and biomarkers of renal function (cystatin C), and inflammation (WBC, CRP-hs, GDF-15, Il-10) (Table S1).

| DISCUSSION
The main findings of the current study were that, in patients with Inflammation is an important factor for the development of atherosclerosis, coronary plaques, and plaque rupture. 1,13,14 A large number of inflammatory markers have been described, and IL-18 has been reported associated with both presence of CV disease and clinical outcomes. 6,7 However, the exact role of these inflammatory mediators in ACS is complex and the importance of IL-18 is only partly understood. The predictive accuracy of these markers can be influenced by the clinical context and the sample time, which is not always well recorded, and consequently the levels of an inflammatory marker prior to an event, including IL-18, is difficult to estimate. 14,15 As might be expected, in the current study of patients with ACS, levels of IL-18 increased from the first measurement at baseline to discharge. 16 Further, the concentrations remained elevated from 1 month until the last observation at 6 months. These findings suggest that the inflammatory response by IL-18 to ACS is rapid but the possible return to pre-ACS levels is slow, as if or when it returns to baseline is currently unknown. In comparison, CRP-hs and WBC have a much faster regression, tapering within days.
The prognostic impact of IL-18 has previously been examined in a smaller cohort (n = 1261) of patients with ACS, which also showed a significant association with long-term (10 years) overall mortality but not with MI. 8 In our study, comprising a more than 10 times larger cohort of patients with ACS, the overall mortality was 5.6% of which the vast majority was CV deaths. An association between the IL-18 levels and CV mortality was also observed and, importantly, the association between IL-18 and CV death remained significant even after the To the left presented overall and to the right in strata by randomized treatment. All subjects with a baseline value and at least one follow-up sample were included in the analysis (n = 4583). The model includes a non-parametric test with the baseline value as a covariate.
adjustment for important CV risk markers as cystatin C, NT-proBNP, and cTnT-hs.
But after the addition of inflammatory biomarkers, this association was no longer statistically significant. This clearly indicates that inflammatory activity is a contributing factor to fatal CV events beyond the We have previously reported that clopidogrel treatment, compared with ticagrelor treatment, was associated with a higher incidence of severe pulmonary infections or sepsis and an increased mortality. 20 This finding was accompanied with a lower leucocyte count at one month and lower mean CRP-hs and IL-6 concentrations in patients randomized to clopidogrel, indicating possible effects of the randomized antiplatelet treatment on immune signaling. 20 However, we found no evidence that the randomized antiplatelet treatment affected IL-18 concentrations. Rather, an enhanced and sustained inflammatory response for more than 6 months after ACS was observed, irrespective of randomized antiplatelet treatment.

| LIMITATIONS
The PLATO trial enrolled a broad ACS population but, as in several clinical trials, certain patient groups were excluded, for example, patients requiring dialysis or patients with recent significant bleeding.
The current substudy investigated IL-18 concentrations at baseline, and all samples were drawn within 24 h of admission, nonetheless, the time span between symptom onset and sample time is not adjusted for and may have affected the observed IL-18 results at baseline. Further, the reported IL-18 concentrations were circulating IL-18, and we were not able to either measure or estimate total IL-18 levels.

| CONCLUSION
In patients with ACS, the IL-18 concentration increased during the first month after the acute event and remained at similar level at 6 months.
The IL-18 levels were consistently associated with male sex, higher BMI, diabetes, decreased renal function, and to other inflammatory biomarkers.
The IL-18 level at baseline was significantly associated with CV-mortality independent of clinical characteristics and indicators of renal and cardiac dysfunction. However, the association with CV mortality was attenuated and did not remain statistically significant after the adjustment for further inflammatory biomarkers and GDF-15. Although the prognostic importance of IL-18 is limited if other prognostic inflammatory biomarkers are available, IL-18 may still be part of underlying inflammatory processes contributing to fatal outcomes in ACS.

ACKNOWLEDGMENTS
The PLATO trial was funded by AstraZeneca. Support for the analysis and interpretation of results and preparation of the manuscript was