Cytokine profiles in the detection of severe lung involvement in hospitalized patients with COVID-19: The IL-8/IL-32 axis

Abstract

Coronavirus disease 2019 (COVID-19) is an infectious respiratory disorder caused by a new coronavirus called SARS-CoV-2. The pathophysiology of severe COVID-19 is associated with a "cytokine storm". IL-32 is a key modulator in the pathogenesis of various clinical conditions and is mostly induced by IL-8. IL-32 modulates important inflammatory pathways (including TNF-α, IL-6 and IL-1b), contributing to the pathogenesis of inflammatory diseases. Il-32 was never evaluated before in COVID-19 patients stratifying as mild-moderate and severe patients. A total of 64 COVID-19 patients, 27 healthy controls were consecutively enrolled in the study. Serum concentrations of biomarkers including IL-1β, IL-10, IFN-γ, TNF-α and IL-6 were quantified by bead-based multiplex analysis and Serum concentration of IL-8 and IL-32 were determined by enzyme-linked immunosorbent assay (ELISA) kits. Interestingly, among the blood parameters, neutrophil and lymphocyte counts were significantly lower in severe COVID-19 patients than in the other, on the contrary, CRP was significantly higher in severe patients than in other groups. The cytokines that best distinguished controls from COVID-19 patients were IL-8 and IL-32, while IL-6 resulted the better variables for discriminate severe group. The best model performance for severe group was obtained by the combination of IL-32, IL-6, IFN-γ, and CRP serum concentration showing an AUC = 0.83. A cut off of 15 pg/ml of IL-6 greatly discriminate survivor from death patients. New insights related to the cytokine storm in COVID-19 patients, highlighting different severity of disease infection.

Keywords: COVID19; Cytokines; Interleukins; Profiles.

Fig. 1

Cytokines concentrations comparison among HC and mild, moderate and severe COVID-19 patients The histograms report mean (center bar) ± SEM (upper and lower bars). If not indicated, p value is not significant. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Fig. 2

a) PCA analysis between HC (orange) and Covid19 patients (blue). b) Heatmap analysis between COVID-19 and HC: The heat map is based on hierarchical clustering applied for row and columns and reported as Euclidean distance. The color of the heat varies from white (indicating relative under-representation) to red (indicating relative over-representation). Clusters are sorted according to adjusted p values, so that the cluster at the top shows the most significant abundance changes between the two conditions (HC and COVID 19 Patients) (c) Decision tree model. The different cytokine levels of patients were employed to create a decision tree model for the detection of best clustering variables.

Fig. 3

a) PCA analysis between Covid19 patients groups: severe, moderate and mild. b) Heatmap analysis between COVID-19 severity groups. The heat map is based on hierarchical clustering applied for row and columns and reported as Euclidean distance. The color of the heat varies from white (indicating relative under-representation) to red (indicating relative over-representation). Clusters are sorted according to adjusted p values, so that the cluster at the top shows the most significant abundance changes between the three conditions (mild, moderate and severe) c) Decision tree model. The different cytokine levels of patients were employed to create a decision tree model for the detection of best clustering variables.

Fig. 4

a). ROC curve analysis of IL-6 (blue), CRP (Red) and INF-γ (green) and IL-32 (black) represent the better combination in discriminate COVID-19 severity group.

Fig. 5

Significant IL-6 values in survival analysis of COVID-19 patients. Kaplan-Maier survival analysis dividing Death Covid-19 patients according to IL-6 (cut-off 15 pg/ml).