P5565S100A8/A9 serum levels as a diagnostic biomarker and its potential connection to NOD2-NLRP3 in patients with a recent onset of myocarditis

Abstract

Abstract Background The alarmin S100A8/A9 has been shown to be of importance in several inflammatory cardiovascular disorders. We recently demonstrated the pivotal role of cardiac S100A8/A9 in human and experimental Coxsackievirus B3 (CVB3)-induced myocarditis (MC). Purpose We aimed to evaluate whether serum S100A8/A9 levels are a marker in patients with a recent onset of MC Methods Serum S100A8/A9, hsCRP, and NT pro-BNP levels were analyzed in patients with a recent onset of MC (≤30 days (d), n=29; ejection fraction (EF): 44.3%±13%), dilated cardiomyopathy patients with inflammation (DCMi: n=112; EF: 28.8%±12%) or without inflammation (DCM: n=58; EF: 26.7%±9%), and controls (co: n=25; EF: 68.5%±5%). Blood samples and endomyocardial biopsies (EMB) were collected at time point (T1). In a subgroup, S100A8/A9 serum levels and EMB were available at T1 (n=10) and follow-up (T2, n=10, mean follow-up 8 months). Results MC ≤30 d patients showed a 4.5-fold (p<0.0001), 19.3-fold (p<0.0001), and 4.0-fold (p<0.0001) increase in S100A8/A9, hsCRP, NT pro-BNP levels vs co, respectively. S100A8/A9 levels correlated with the disease activity, displayed by EMB counts of inflammatory cells (CD3: r=0.464, p=0.0128, XY pairs=28, LFA-1: r=0.551, p=0.002, XY pairs=28, Mac-1: r=0.418, p=0.026, XY pairs=28), and the EF (r=0.545, p=0.0027, XY pairs=28). MC ≤30 d patients showed an association between serum S100A8/A9 levels and EMB S100A8 (r=0.482, p=0.060, XY pairs=16), S100A9 (r=0.441, p=0.088, XY pairs=16), nucleotide-binding oligomerization domain containing-protein 2 (NOD2, r=0.55, p=0.035, XY=15), and Nod-like receptor family, pyrin domain-containing 3 protein (NLRP3, r=0.52, p=0.048, XY=15) mRNA levels. Also EMB S100A8 and S100A9 mRNA levels showed a significant correlation with EMB NOD2 and NLRP3 mRNA expression. Serum S100A8/A9 levels were increased by 3.0-fold (p<0.0001) and 1.8-fold (p=0.0005) in DCMi (n=112), and DCM (n=58) patients vs co, respectively. However, the S100A8/A9 levels of DCMi and DCM patients were 1.5-fold (p=0.07) and 2.5-fold (p<0.0001) lower vs MC ≤30 d patients. ROC analyses of S100A8/A9 in MC ≤30 d provided a cut-off of 583 ng/ml with a specificity=92%, sensitivity=86.2%, a PPV=92.6%, a NPV=85.2%, and an AUC=0.934 vs co, which was superior to hsCRP (cut-off=5 mg/l): specificity=95.8%, sensitivity=58.6%, a PPV=94.4%, a NPV=65.7%, AUC=0.885. In the subgroup, S100A8/A9 levels decreased after heart failure medication (T1: 2454±1931 ng/ml vs T2: 934.4±552 ng/ml; p=0.002), reflected by a decrease of EMB inflammatory cells. Baseline serum S100A8/A9 levels predicted the change in EMB CD3 and Mac-1. Conclusions These results support an additional value for S100A8/A9 serum levels as a potential diagnostic biomarker and as a tool to monitor the course of the disease. We provide first evidence that S100A8/A9 is connected to the NOD2-NLRP3 pathway in these patients. Acknowledgement/Funding Novartis