Proteomic-based neutrophil phenotype characterization identifies patients with increased risk of hospitalization due to heart failure

E. Haj-Yehia (Essen)1, S. Gosh (Dortmund)2, P. Schulte (Essen)1, S. Korste (Essen)1, A.-A. Mahabadi (Essen)1, M. Gunzer (Essen)3, A. Sickmann (Dortmund)4, U. Hendgen-Cotta (Essen)1, T. Rassaf (Essen)1, M. Totzeck (Essen)1
1Universitätsklinikum Essen Klinik für Kardiologie und Angiologie Essen, Deutschland; 2Dortmund, Deutschland; 3Universität Duisburg-Essen Institut für experimentelle Immunologie und Bildgebung Essen, Deutschland; 4Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V. Dortmund, Deutschland

Aims

Neutrophils play an important role in acute and chronic cardiovascular diseases. Some neutrophil properties support tissue healing, while others provoke additional injury e.g. through release of neutrophil granule peptides and reactive oxygen species (ROS). The impact of this phenotypic spectrum of neutrophil functions on patient outcome and whether these can be used as tractable therapeutic targets remains unclear. Therefore, this study aimed to characterize the individual neutrophil phenotype with the help of proteomic analysis in patients with cardiovascular disease to identify those phenotypes associated with major cardiac outcome events.  

Methods and results

A population of 215 out of 2,845 patients (median age 72 years (interquartile range (IQR): 61-80 years); 33.0% female) from the prospective all-comer Essen Registry of Coronary Artery Disease II (ECAD II) were analyzed. Among other cardiovascular diseases, 113 patients (52.6%) exhibited coronary artery disease, 21 patients (9.8%) history of myocardial infarction and 109 patients (50.7%) heart failure. For included patients, untargeted proteomic measurements of fresh isolated neutrophils from peripheral blood at registry enrollment were provided. Based on the proteome analysis, we identified three distinct patient clusters using uniform manifold approximation and projection (UMAP). 774 out of 4,474 identified proteins were found in all three cluster with 600 proteins showing statistical significance. During follow-up (median follow-up time: 285 days (IQR: 224-346 days)), patient clusters with higher expression levels of pro-inflammatory granule peptides and decreased expression of enzymes associated with ROS detoxification showed significantly increased rates of hospitalization for heart failure (hazard ratio (HR): 5.92; 95% confidence interval (CI): 1.67-21.06; p = 0.01), while no differences were observed concerning occurrence of major adverse cardiac events (MACE) consisting of non-fatal myocardial infarction, non-fatal stroke and cardiac death. The association for hospitalization for heart failure was unaffected by adjustment for further confounders as age, sex and concomitant cardiovascular diseases (HR: 5.04; 95% CI: 1.12-22.62; p = 0.04). 

Conclusions

To our knowledge, this study is the first approach using proteomic phenotyping to analyze the impact of neutrophil functional alterations on clinical outcome in patients with cardiovascular disease. Besides increased expression of pro-inflammatory neutrophil granule peptides, reduction in proteins associated with ROS detoxification were also associated with increased rates of hospitalization for heart failure. Especially the latter indicates a novel involvement of neutrophils during cardiac remodeling, which needs further investigations in future studies.