Plasma alarmin S100A8/A9 serves as a potential biomarker of major adverse cardiovascular events after acute myocardial infarction

Introduction

Despite advances in primary interventional therapies and pharmacologic treatments, the residual risk of ischemic cardiovascular events following acute myocardial infarction (AMI) persists at elevated levels (1-3). There remains a critical need to identify high-risk patients prone to recurrent events or premature mortality and to optimize therapeutic strategies accordingly. Consequently, novel circulating biomarkers reflecting distinct pathophysiological pathways in acute ischemic heart disease have emerged as promising tools for risk stratification, enabling targeted preventive interventions in vulnerable subgroups (4,5).

S100A8/A9, a calcium-binding heterodimer of the S100 protein family, has garnered significant interest as a pivotal alarm in that it amplifies inflammatory cascades upon release (6). Preclinical and clinical studies have established a pathophysiological connection between S100A8/A9 and myocardial ischemia. Its expression is markedly upregulated in both infarcted myocardial tissue and circulating blood of AMI patients (7). Notably, elevated serum S100A8/A9 levels during AMI are not cardiomyocyte-derived, implicating inflammatory cells as the primary source (8). Further mechanistic investigations revealed that S100A8/A9 exacerbates post-MI heart failure (HF) by driving nuclear factor-κB-mediated proinflammatory cytokine production (9).

Over the past decade, accumulating evidence has positioned S100A8/A9 as a potential predictive biomarker of cardiovascular diseases (10). Elevated serum S100A8/A9 levels have been reported in acute coronary syndrome patients and proposed its utility for cardiovascular risk stratification and outcome prediction (11). However, these findings were constrained by limited sample sizes. To address this gap, we conducted a large-scale prospective cohort study to evaluate the prognostic value of plasma S100A8/A9 levels in hospitalized AMI patients. We present this article in accordance with the STARD reporting checklist (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-1-626/rc).

Methods

Study design and population

Between August 2013 and June 2016, a total of 1,312 consecutive patients hospitalized with ST-segment elevation myocardial infarction (STEMI) or non-ST-segment elevation myocardial infarction (NSTEMI) in Shanghai East Hospital, Tongji University (Shanghai, China), were prospectively enrolled in this cohort study. Participants were required to present cardiac troponin I (TnI) elevation exceeding the 99^th percentile of the upper reference within the 24 hours of admission, and meet at least one of the following: (I) chest pain lasting over 20 minutes; (II) diagnostic serial electrocardiographic changes indicative of incident ischemia; (III) other imaging or pathological evidence of myocardial necrosis. Patients with acute infection, autoimmune diseases, known malignancy, or end-stage renal disease were excluded from our study. The time from symptom onset to admission had to be within 48 hours.

Baseline covariates including demographic characteristics, comorbidity status (hypertension, diabetes mellitus, hyperlipidemia), smoking status, body mass index (BMI), number of vessels involved, previous treatment history, prior MI, Killip class, pharmacological treatments, and levels of TnI and brain natriuretic peptide (BNP) were collected through electronic medical records and in-person interviews. Smoking was defined as regular tobacco use within the past year. The left ventricular ejection fraction (LVEF) was visually estimated by experienced cardiologists using echocardiography, guided by strain analysis and wall motion index. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the ethics committee of Shanghai East Hospital (No. 2025-DF-01574), and written informed consent was obtained from all participants.

Outcome definition and follow-up

The primary endpoint was the time to major adverse cardiovascular events (MACEs), a composite endpoint defined as the first occurrence of any of the following: all-cause mortality, hospitalization for HF, or recurrent MI. Each component of MACE was also analyzed as a secondary endpoint. All-cause mortality was defined as any death occurring during the follow-up period. Hospitalization for HF was defined as unplanned readmission requiring management of decompensated HF with high-dose intravenous diuretics, inotropic agents, or intravenous nitrates, as explicitly documented in medical records. Recurrent MI was defined according to the Fourth Universal Definition of Myocardial Infarction, requiring elevation of cardiac troponin values above the 99th percentile upper reference limit plus at least one of the following: symptoms of myocardial ischemia, new ischemic electrocardiographic changes, development of pathological Q waves, or imaging evidence of new loss of viable myocardium or regional wall motion abnormality. Trained staff prospectively entered data into a central database. Follow-up outcomes at 24 months were ascertained via hospital records and telephone contact, then verified against medical charts.

Laboratory methods

Blood samples were collected in EDTA tubes upon admission. Plasma was separated by centrifugation at 3,000 ×g for 10 minutes and stored at −80 ℃ until analysis. Plasma S100A8/A9 concentrations were quantified using a commercially available human S100A8/A9 ELISA kit (Sigma-Aldrich, St. Louis, MO, USA). Duplicate samples were analyzed to ensure reliable results.

Sensitivity analyses

To further address potential residual confounding related to acute-phase clinical instability and reverse causation, sensitivity analyses were performed by excluding patients who experienced study outcomes within the first 6 months after index admission. Cox proportional hazards models were refitted using the same multivariable adjustment strategy as in the primary analyses.

Statistical analysis

Medians [interquartile range (IQR)] were used to describe continuous variables, while frequencies and percentages were used for categorical variables. For statistical comparison between groups, the Mann-Whitney U test was applied to continuous data, and the Chi-squared test was utilized for categorical data.

Cumulative incidences were estimated via Kaplan-Meier analysis for MACE and all-cause mortality, with comparisons between S100A8/A9 tertiles performed using the log-rank test. While for recurrent MI and hospitalization due to HF, cumulative incidences were compared using Finn-Gray models with death and another non-fatal component of MACE as competing events. Cox proportional hazards models were employed to assess associations between S100A8/A9 tertiles and outcomes, adjusting for age at recruitment, sex, comorbidities (hypertension, diabetes, hyperlipidemia), smoking status, BMI, multivessel disease, revascularization [percutaneous coronary intervention (PCI)/coronary artery bypass grafting (CABG)], prior MI, LVEF, Killip class, log-transformed peak TnI and BNP levels, and any guideline-directed medical therapies [including dual antiplatelet therapy, diuretic, statin, B-blockers, angiotensin-converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB), and nitrate]. The proportional hazards assumption was validated using Schoenfeld residuals, and hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) are provided. Nonlinear associations between plasma S100A8/A9 concentrations and clinical outcomes were modeled using restricted cubic splines.

To assess model stability, the number of events per variable (EPV) was calculated for each fully adjusted Cox proportional hazards model. Potential multicollinearity between S100A8/A9 and established markers of infarct severity (TnI, BNP, LVEF, and Killip class) was evaluated using Pearson correlation coefficients and variance inflation factors (VIFs). Internal validation was performed using bootstrap resampling (1,000 repetitions) to assess stability of HR estimates and model optimism. Optimism-corrected C-indices were derived from bootstrap validation. Given the relatively low EPVs for all-cause mortality, HF hospitalization, and recurrent MI, Firth’s penalized Cox regression was conducted as a sensitivity analysis to address potential sparse-event bias and separation.

The reference model included age; gender; hypertension; diabetes mellitus; hyperlipidemia; smoking; BMI; multivessel disease; PCI; CABG; previous MI; LVEF; Killip class; log-transformed peak TnI and BNP. Discrimination for time-to-event outcomes was evaluated using Harrell’s C-index derived from Cox proportional hazards models, which account for censoring. To formally assess improvement in discrimination after addition of S100A8/A9, bootstrap resampling (1,000 repetitions) was performed. In each bootstrap sample, both models were refitted and the difference in C-index (ΔC-index) was calculated. Empirical 95% CIs were derived from the bootstrap distribution. Model optimism was assessed using bootstrap validation, and optimism-corrected C-indices were reported. Calibration was evaluated using bootstrap-corrected calibration plots based on Cox models at a prespecified time horizon (2.93 years).

Statistical processing was executed using R version 3.4 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Patient characteristics

This study enrolled 1,312 AMI patients, comprising both STEMI (58.5%) and NSTEMI (41.5%) subtypes, with the baseline characteristics of participants summarized in Table 1. The cohort demonstrated a male predominance (64.6%, n=847) with median age 66 years (IQR, 57–76 years). Stratification by plasma S100A8/A9 tertiles revealed progressive clinical severity across biomarker levels: Tertile 1 (≤4,259.0 pg/mL), Tertile 2 (4,259.1–5,650.6 pg/mL), and Tertile 3 (>5,650.6 pg/mL). Compared with patients with lower S100A8/A9 levels, patients with higher S100A8/A9 levels exhibited significantly higher levels of LDL (P=0.03) and CRP (P=0.008), higher prevalence of prior MI (P<0.001), more severe left ventricular dysfunction (P<0.001), and higher Killip class (P<0.001). Notably, escalation of S100A8/A9 correlated with elevated myocardial injury markers (peak TnI and BNP, both P<0.001). Baseline demographics including age, sex distribution, and comorbidity profiles showed no significant differences (all P>0.05).

Association of S100A8/A9 with prognosis

Over a follow-up period of 2.93 years, 132 patients died (incidence rate: 10.1%), and 392 patients (incidence rate: 29.9%) reached the composite MACE endpoint. S100A8/A9 concentrations were significantly elevated in patients who experienced MACE compared to those who did not [8,976 pg/mL (IQR, 5,821–11,219 pg/mL) vs. 4,332 pg/mL (IQR, 3,412–5,281 pg/mL); P<0.001].

Restricted cubic spline analyses revealed significant non-linear associations between plasma S100A8/A9 concentrations and all clinical outcomes (all P overall <0.001; Figure S1). For MACE, the HR crossed 1 at approximately 5.18 ng/mL (≈5,184 pg/mL), followed by a progressive increase in risk with higher S100A8/A9 levels. Similarly, for recurrent MI, the HR crossed 1 at approximately 4.81 ng/mL (≈4,810 pg/mL), with a steep rise in risk thereafter. For HF hospitalization, the HR crossed 1 at approximately 7.38 ng/mL (≈7,381 pg/mL), indicating a higher S100A8/A9 concentration associated with increased HF risk. In contrast, the association with all-cause mortality was non-monotonic, with the HR peaking at approximately 6.45 ng/mL (≈6,446 pg/mL) before declining at higher concentrations. These findings indicate distinct, outcome-specific patterns in the relationship between S100A8/A9 and long-term cardiovascular outcomes. Kaplan-Meier analysis demonstrated that patients with higher tertile of S100A8/A9 exhibited significantly increased risks of MACE and all-cause mortality (both P<0.001) (Figure 1). Competing risk analyses using Finn-Gray models further revealed that elevated S100A8/A9 predicted higher cumulative incidence of rehospitalization due to HF and recurrent MI treating death and another non-fatal component of MACE as competing events (both P<0.001) (Figure S2).

Figure 1 Cumulative incidence of MACE and all-cause mortality according to S100A8/A9 tertiles. The cumulative incidence of MACE (A) and all-cause mortality (B). Log-rank tests were used to calculate P values. MACE, major adverse cardiac event.

Univariable Cox proportional hazards analysis showed a higher risk of MACE (HR =4.78; 95% CI: 4.00–5.71) and all-cause mortality (HR =1.69; 95% CI: 1.35–2.11) in patients with increased circulating level of S100A8/A9 during long-term follow-up (P_trend<0.001) (Table 2). The association of S100A8/A9 concentrations with MACE (HR =3.75; 95% CI: 3.05–4.62) and all-cause mortality (HR =1.68; 95% CI: 1.35–2.10) remains significant after full multivariable adjustment (P_trend<0.001) (Table 2). Furthermore, the higher level of S100A8/A9 was also associated with hospitalization due to HF and recurrent MI in unadjusted model, the association also persisted after full adjustment (P_trend<0.001) (Table 2). Sensitivity analysis was shown that after adjusting for baseline CRP level, the association between S100A8/A9 and MACE remained significant. After further excluding subjects with MACE within the first 30 days, the result remained consistent.

Sensitivity analyses to further assess the robustness of these findings and reduce potential confounding from early clinical instability, sensitivity analyses were performed after excluding patients who experienced outcomes within the first 6 months after admission (Table S1). After exclusion of early events, elevated S100A8/A9 levels remained consistently associated with increased risks of MACE, HF hospitalization, and recurrent MI. For all-cause mortality, effect estimates were modestly attenuated after exclusion of early events but remained significant in the highest tertile, and the overall trend remained statistically significant.

For the primary endpoint (MACE), EPV was 17.81 (392 events, 22 covariates), indicating adequate information density and acceptable model stability. For secondary endpoints (all-cause mortality, HF hospitalization, and recurrent MI), EPV ranged from 6.0 to 7.04, reflecting relatively lower event density and potentially greater estimate variability. Pearson correlation coefficients and VIFs were calculated to evaluate potential multicollinearity between S100A8/A9 and established infarct severity markers (TnI, BNP, LVEF, and Killip class). All pairwise correlations were moderate (|r|<0.6). VIF values ranged from 1.18 to 2.97, well below conventional thresholds for concern (VIF >5). These findings suggest that S100A8/A9 provides non-redundant prognostic information independent of traditional markers. Internal validation was performed using bootstrap resampling (1,000 repetitions) (Table S2). For MACE, the apparent C-index was 0.801 and the optimism-corrected C-index was 0.791 (optimism 0.010), indicating minimal overfitting. The bootstrap median HR for Tertile 3 vs. Tertile 1 was 9.73 (95% bootstrap interval: 6.25–15.84), closely aligned with the primary estimate (HR =9.76), supporting model robustness. For HF hospitalization and recurrent MI, bootstrap HR estimates remained directionally consistent with the primary models. For all-cause mortality, bootstrap estimates were stable and similar to the original model [bootstrap HR 3.19 (2.04–5.72)], suggesting that the association is not attributable to model instability despite lower EPV. Given the relatively lower EPV for secondary endpoints, Firth’s penalized Cox regression was performed as a sensitivity analysis to address potential sparse-event bias and separation (Table S3). Penalized estimates remained directionally consistent with the primary Cox models. These results indicate that the large HRs observed for HF hospitalization and recurrent MI are unlikely to be solely driven by separation or sparse-event bias.

Prognostic value of S100A8/A9

Using Cox model-based Harrell’s C-index, discrimination for MACE improved from 0.752 in the reference model to 0.822 after addition of log-transformed S100A8/A9 (Table 3). Bootstrap resampling demonstrated a median ΔC-index of 0.070 (95% CI: 0.051–0.090; P<0.001), confirming a robust improvement in discrimination. In contrast, for all-cause mortality, the addition of S100A8/A9 resulted in minimal change in C-index (0.718 to 0.720), with bootstrap ΔC-index 0.002 (95% CI: −0.005, 0.013; P=0.58), indicating no significant incremental discrimination.

Bootstrap-corrected calibration plots demonstrated good agreement between predicted and observed risks for both MACE and all-cause mortality (Figure S3). Optimism-corrected curves closely overlapped with apparent curves, suggesting minimal overfitting. Minor deviation was observed in the extreme high-risk range for MACE, likely reflecting limited sample size in this stratum. Addition of S100A8/A9 did not materially impair calibration.

Discussion

We investigated the prognostic utility of S100A8/A9 levels regarding MACE and its subtypes in AMI patients. Long-term analysis revealed that S100A8/A9 levels were significantly linked to MACE, all-cause mortality, recurrent MI, and HF hospitalization, independent of standard risk factors. Furthermore, adding S100A8/A9 to the reference model significantly enhanced the prediction of MACE, suggesting that it serves as a valuable novel biomarker for risk stratification in post-AMI patients.

S100A8/A9 (also known as myeloid-related protein MRP-8 and MRP-14) is an endogenous damage-associated molecular pattern (DAMP) or alarmin of the S100 family, encoded by a gene cluster on chromosome 1q21 (12,13). Numerous other DAMPs, including high mobility group box-1 (HMGB1), ATP, uric acid, fibronectin extra domain A, interleukin-1β and heat shock proteins, are also involved in the post-MI inflammatory response (14). Among these, S100A8/A9 constitutes up to 45% of cytosolic proteins in neutrophils, the cells that dominate the early immune response post-MI by recruiting other immune cells and releasing inflammatory mediators (15).

Interestingly, the initial wave of infiltrating neutrophils appears to drive inflammation amplification through the massive release of S100A8/A9. Acting in an autocrine or paracrine manner, extracellular S100A8/A9 interacts with TLR4 on naïve neutrophils, priming the NLRP3 inflammasome and promoting IL-1β secretion (16). Disruption of this S100A8/A9-NLRP3-IL-1β axis via genetic or pharmacological strategies has been shown to dampen myelopoiesis and improve cardiac function (11,17). These findings suggest that during MI and potentially other ischemic injuries, neutrophils regulate the production of leukocytes by deploying S100A8/A9. Additionally, S100A8/A9 released by neutrophils may link to poor clinical outcomes by causing irreversible damage to mitochondrial electron transport chain complex 1 (ETC1), leading to cardiomyocyte apoptosis (11,18).

S100A8/A9 (MRP-8/14) is an endogenous DAMP in the S100 family, located on chromosome 1q21. It plays a key role in the post-MI inflammatory response alongside other DAMPs like HMGB1, ATP, and uric acid. S100A8/A9 is highly abundant in neutrophils, comprising nearly 45% of their cytosolic protein, and is crucial for the early immune response following MI.

Infiltrating neutrophils amplify inflammation by releasing S100A8/A9, which then binds to TLR4 on other neutrophils to prime the NLRP3 inflammasome and trigger IL-1 secretion. Strategies that disrupt this signaling axis have successfully reduced myelopoiesis and improved cardiac function. This suggests neutrophils use S100A8/A9 to self-regulate leukocyte production during ischemic injury. Furthermore, S100A8/A9 may worsen outcomes by damaging mitochondrial ETC1, thereby inducing cardiomyocyte apoptosis.

The large effect sizes observed for S100A8/A9 warrants careful interpretation. We acknowledge that S100A8/A9 is closely linked to baseline disease severity and systemic inflammatory activation in the AMI setting. However, several observations argue against these associations being explained solely by residual confounding.

First, extensive multivariable adjustment was performed, incorporating established markers of infarct severity, hemodynamic instability, and cardiac dysfunction, including peak TnI, LVEF, BNP, and Killip class, as well as coronary anatomy, revascularization strategy, and guideline-directed medical therapy. Second, sensitivity analyses excluding events occurring within the first 6 months yielded consistent results, indicating that the associations are not primarily driven by early clinical instability or reverse causation.

Although unmeasured confounding cannot be fully excluded in an observational study, the persistence of strong associations after adjustment for all routinely available indicators of clinical instability suggests that S100A8/A9 captures prognostic information beyond conventional severity markers. Rather than representing an isolated causal factor, S100A8/A9 likely functions as an integrative biomarker reflecting sustained inflammatory vulnerability and susceptibility to adverse cardiovascular events following AMI.

The spline analyses provide important insight into the clinical interpretation of S100A8/A9. Although several reference points could be identified—such as the HR =1 crossings for MACE, recurrent MI, and HF hospitalization, and the peak risk for all-cause mortality—these values differed substantially across outcomes. This heterogeneity suggests that S100A8/A9 does not operate through a single universal risk threshold but rather reflects endpoint-specific pathophysiological processes.

Importantly, none of the spline curves demonstrated a sharp inflection point indicating an abrupt transition in risk. Instead, risk changed gradually across the observed range of S100A8/A9 concentrations, supporting the interpretation of S100A8/A9 as a continuous risk marker rather than a dichotomous classifier. Although receiver-operating characteristic (ROC) analyses confirmed the strong discriminatory ability of S100A8/A9, particularly for MACE, ROC-derived cutoffs are inherently dependent on outcome prevalence and do not account for non-linear or non-monotonic risk patterns. In the absence of clearly defined intervention strategies or external validation, selecting a single ROC-based cutoff could therefore be arbitrary and potentially misleading. Accordingly, the spline-derived values should be viewed as exploratory, outcome-specific reference points rather than definitive clinical decision thresholds. Furthermore, we revealed an incremental value of S100A8/A9 over the reference model combining clinical risk factors and serum biomarkers in the prediction of MACE but not all-cause mortality. The current circulating biomarkers including TnI and BNP enabling risk stratification for AMI can provide prognostic value for MACE in patients with AMI (19-21). Consistent with previous studies, BNP and TnI were also strongly associated with the future risk of adverse events in this study_. Compared to established cardiac biomarkers such as TnI and BNP, we found that S100A8/A9 has a higher prognostic ability for MACE. Complementary methods including calibration, discrimination, and reclassification analyses were also used to confirm the incremental prognostic value of S100A8/A9. Additionally, to account for potential acute-phase confounding factors such as baseline disease severity and systemic inflammatory burden, sensitivity analyses were conducted. These analyses excluded early events occurring within the first 30 days and incorporated additional adjustment for C-reactive protein (CRP). Remarkably, the association between S100A8/A9 and long-term MACEs remained robustly significant, further underscoring its independent prognostic utility. Recently, a similar study revealed that S100A8/A9 was independently associated with higher risk of HF after AMI, with better predictive performance than traditional biomarkers (22). Our study further expands the potential applications of S100A8/A9 as a predictor of adverse outcomes after AMI.

Aforementioned dual mechanism of S100A8/A9 may explain its superior MACE prediction, as composite endpoints capture both inflammatory complications and ischemic recurrences. However, mortality determinants incorporate a wide range of systemic factors with varying pathophysiologic links to neutrophil activity, in which the role of S100A8/A9 may be diluted by other irrelevant biological mechanisms. The differential performance was also probably related to distinct biomarker kinetics. S100A8/A9 elevation mirrors acute neutrophil activity during early post-AMI inflammation, which acts as a critical window for MACE initiation, while TnI and BNP capture cumulative myocardial injury progressing throughout recovery. Mortality risk, however, integrates both acute-phase events and later chronic complications influenced by comorbidities and is less correlated with initial inflammatory surges. Our single-timepoint measurement at admission may thus preferentially capture MACE-related pathways.

There are some limitations in the present study. Firstly, S100A8/A9 was measured only once at admission. Given the dynamic inflammatory response after AMI, a single measurement may not adequately capture the biomarker’s prognostic trajectory. Besides, the kinetics of S100A8/A9 in the post-AMI setting remain less well-characterized. Serial measurements would be valuable to determine the optimal timing of sample collection for predicting prognosis and to understand how S100A8/A9 levels evolve relative to other biomarkers over time. Nevertheless, well-characterized inflammatory biomarkers like CRP are routinely measured at admission in clinical practice, capturing the peak systemic inflammatory response during AMI’s early phase. Secondly, our study was conducted in a single center with an exclusively Chinese population, which may limit the generalizability of our findings to other ethnic groups. Ethnic differences in inflammatory response regulation—including genetic polymorphisms affecting S100A8/A9 expression, clearance, and receptor binding affinity—could alter the biomarker’s prognostic performance across populations. Additionally, cardiovascular risk factor profiles (e.g., prevalence of hypertension, diabetes mellitus, and lipid disorders) vary significantly between Chinese and Western populations, which may affect the association between S100A8/A9 levels and adverse outcomes. Future multi-center, multi-ethnic studies are warranted to validate the prognostic utility of S100A8/A9 across diverse populations and healthcare contexts (23,24). Thirdly, lack of harmonized measurement protocols across centers may delay clinical implementation of S100A8/A9 as a prognostic biomarker of AMI. Finally, further prospective studies on the clinical effectiveness of using this biomarker to distinguish low- or high-risk groups and to choose strategies need to be performed to improve the management system of patients after AMI.

The magnitude of HRs observed for MACE and its cardiovascular components (HF hospitalization and recurrent MI) is larger than typically reported for single circulating biomarkers in AMI populations. However, several analyses support the robustness of these estimates. First, multicollinearity analyses demonstrated that S100A8/A9 was not strongly correlated with established severity markers, indicating that the observed associations are unlikely to reflect redundancy with traditional indicators of infarct burden. Second, bootstrap resampling showed stable HR estimates with minimal optimism in discrimination, supporting model robustness. Third, Firth’s penalized Cox regression yielded directionally consistent results for endpoints with lower EPV, suggesting that sparse-event bias or separation does not fully explain the magnitude of the associations.

Importantly, the large HRs were primarily driven by event concentration within the highest tertile (T3) of S100A8/A9. Comparisons between T2 and T1 demonstrated substantially smaller effect sizes across endpoints. When S100A8/A9 was modeled as a log-transformed continuous variable, HRs were more moderate, indicating that tertile-based categorization accentuates contrasts between extreme inflammatory strata. These findings suggest that the highest tertile likely represents a subgroup with markedly elevated inflammatory burden and clinical vulnerability rather than reflecting generalized instability of the regression model. While this paper robustly establishes the statistical prognostic value of S100A8/A9 for MACE, a DCA would be essential to evaluate its net clinical benefit and move the findings closer to actual clinical implementation.

S100A8/A9 was measured only once at admission. We acknowledge that the limited understanding of S100A8/A9 kinetics in the post-AMI setting is an important limitation, especially when compared to biomarkers with well-characterized temporal profiles such as cardiac troponin TNI and BNP. Nevertheless, well-characterized inflammatory biomarkers like CRP are routinely measured at admission in clinical practice, capturing the peak systemic inflammatory response during AMI’s early phase (23,24). Serial measurements in future studies will be valuable to further characterize the full kinetic trajectory of S100A8/A9 post-AMI, identify the optimal sampling time point for prognostic stratification, and compare its dynamic changes with those of TnI and BNP across the post-MI recovery period. In summary, while the single admission measurement of S100A8/A9 appears to effectively capture the early inflammatory vulnerability crucial for MACE prediction, its prognostic reach might be limited for broader outcomes like all-cause mortality due to the lack of information on its dynamic changes over.

Conclusions

In conclusion, the present study demonstrates that elevated circulating S100A8/A9 levels are associated with higher risks of adverse outcomes in post-AMI patients. Notably, the prognostic value of S100A8/A9 is significantly more pronounced for composite cardiovascular outcomes (MACE)—particularly recurrent MI and HF hospitalization—than for all-cause mortality. Future prospective studies are warranted to further validate the prognostic utility of S100A8/A9 in clinical practice.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-1-626/rc

Data Sharing Statement: Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-1-626/dss

Peer Review File: Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-1-626/prf

Funding: This study was supported by grants from the National Natural Science Foundation of China (No. 82270432), Science Foundation of Shanghai Pudong Municipal Health Commission (No. PW2022A-06), Young Medical Talents Training Program of Health Bureau of Shanghai (No. 2022YQ035), Key Disciplines Group Construction Project of Shanghai Pudong New Area Health Commission (No. PWZxq2022-02), Research Program of Shenzhen Science and Technology Innovation Committee (No. JCYJ20220530163603008), Natural Science Foundation of Guangdong Province (Nos. 2023A1515010880 and 2025A1515012052), and the Key Research Center Construction Project of Shanghai (No. 2022ZZ01008).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-1-626/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the ethics committee of Shanghai East Hospital (No. 2025-DF-01574), and written informed consent was obtained from all participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Roe MT, Armstrong PW, Fox KA, et al. Prasugrel versus clopidogrel for acute coronary syndromes without revascularization. N Engl J Med 2012;367:1297-309. [Crossref] [PubMed]
2. de Lemos JA, Morrow DA, Bentley JH, et al. The prognostic value of B-type natriuretic peptide in patients with acute coronary syndromes. N Engl J Med 2001;345:1014-21. [Crossref] [PubMed]
3. Jernberg T, Hasvold P, Henriksson M, et al. Cardiovascular risk in post-myocardial infarction patients: nationwide real world data demonstrate the importance of a long-term perspective. Eur Heart J 2015;36:1163-70. [Crossref] [PubMed]
4. Zhelev Z, Hyde C, Youngman E, et al. Diagnostic accuracy of single baseline measurement of Elecsys Troponin T high-sensitive assay for diagnosis of acute myocardial infarction in emergency department: systematic review and meta-analysis. BMJ 2015;350:h15. [Crossref] [PubMed]
5. Chapman AR, Anand A, Boeddinghaus J, et al. Comparison of the Efficacy and Safety of Early Rule-Out Pathways for Acute Myocardial Infarction. Circulation 2017;135:1586-96. [Crossref] [PubMed]
6. Pruenster M, Vogl T, Roth J, et al. S100A8/A9: From basic science to clinical application. Pharmacol Ther 2016;167:120-31. [Crossref] [PubMed]
7. Katashima T, Naruko T, Terasaki F, et al. Enhanced expression of the S100A8/A9 complex in acute myocardial infarction patients. Circ J 2010;74:741-8. [Crossref] [PubMed]
8. Du CQ, Yang L, Han J, et al. The elevated serum S100A8/A9 during acute myocardial infarction is not of cardiac myocyte origin. Inflammation 2012;35:787-96. [Crossref] [PubMed]
9. Volz HC, Laohachewin D, Seidel C, et al. S100A8/A9 aggravates post-ischemic heart failure through activation of RAGE-dependent NF-κB signaling. Basic Res Cardiol 2012;107:250. [Crossref] [PubMed]
10. Cotoi OS, Dunér P, Ko N, et al. Plasma S100A8/A9 correlates with blood neutrophil counts, traditional risk factors, and cardiovascular disease in middle-aged healthy individuals. Arterioscler Thromb Vasc Biol 2014;34:202-10. [Crossref] [PubMed]
11. Marinković G, Grauen Larsen H, Yndigegn T, et al. Inhibition of pro-inflammatory myeloid cell responses by short-term S100A9 blockade improves cardiac function after myocardial infarction. Eur Heart J 2019;40:2713-23. [Crossref] [PubMed]
12. Sreejit G, Flynn MC, Patil M, et al. S100 family proteins in inflammation and beyond. Adv Clin Chem 2020;98:173-231. [Crossref] [PubMed]
13. Lim SY, Raftery MJ, Geczy CL. Oxidative modifications of DAMPs suppress inflammation: the case for S100A8 and S100A9. Antioxid Redox Signal 2011;15:2235-48. [Crossref] [PubMed]
14. Prabhu SD, Frangogiannis NG. The Biological Basis for Cardiac Repair After Myocardial Infarction: From Inflammation to Fibrosis. Circ Res 2016;119:91-112. [Crossref] [PubMed]
15. Sreejit G, Abdel-Latif A, Athmanathan B, et al. Neutrophil-Derived S100A8/A9 Amplify Granulopoiesis After Myocardial Infarction. Circulation 2020;141:1080-94. [Crossref] [PubMed]
16. Nagareddy PR, Kraakman M, Masters SL, et al. Adipose tissue macrophages promote myelopoiesis and monocytosis in obesity. Cell Metab 2014;19:821-35. [Crossref] [PubMed]
17. Sager HB, Heidt T, Hulsmans M, et al. Targeting Interleukin-1β Reduces Leukocyte Production After Acute Myocardial Infarction. Circulation 2015;132:1880-90. [Crossref] [PubMed]
18. Wang Y, Shi Y, Shao Y, et al. S100A8/A9(hi) neutrophils induce mitochondrial dysfunction and PANoptosis in endothelial cells via mitochondrial complex I deficiency during sepsis. Cell Death Dis 2024;15:462. [Crossref] [PubMed]
19. Aleksova A, Fluca AL, Beltrami AP, et al. Biomarkers of Importance in Monitoring Heart Condition After Acute Myocardial Infarction. J Clin Med 2024;14:129. [Crossref] [PubMed]
20. Aleksova A, Sinagra G, Beltrami AP, et al. Biomarkers in the management of acute heart failure: state of the art and role in COVID-19 era. ESC Heart Fail 2021;8:4465-83. [Crossref] [PubMed]
21. Apple FS, Collinson POIFCC Task Force on Clinical Applications of Cardiac Biomarkers. Analytical characteristics of high-sensitivity cardiac troponin assays. Clin Chem 2012;58:54-61. [Crossref] [PubMed]
22. Ma J, Li Y, Li P, et al. S100A8/A9 as a prognostic biomarker with causal effects for post-acute myocardial infarction heart failure. Nat Commun 2024;15:2701. [Crossref] [PubMed]
23. Deftereos S, Giannopoulos G, Angelidis C, et al. Anti-Inflammatory Treatment With Colchicine in Acute Myocardial Infarction: A Pilot Study. Circulation 2015;132:1395-403. [Crossref] [PubMed]
24. Li Y, He S, Liu T, et al. Effect of high-sensitivity C-reactive protein on the relationship between haemoglobin A1c and cardiovascular events in patients with acute coronary syndrome undergoing percutaneous coronary intervention: a cohort study. Cardiovasc Diagn Ther 2022;12:614-25. [Crossref] [PubMed]