Evolution of NETosis markers and DAMPs have prognostic value in critically ill COVID-19 patients

Abstract

Coronavirus disease 19 (COVID-19) presents with disease severities of varying degree. In its most severe form, infection may lead to respiratory failure and multi-organ dysfunction. Here we study the levels and evolution of the damage associated molecular patterns (DAMPS) cell free DNA (cfDNA), extracellular histone H3 (H3) and neutrophil elastase (NE), and the immune modulators GAS6 and AXL in relation to clinical parameters, ICU scoring systems and mortality in patients (n = 100) with severe COVID-19. cfDNA, H3, NE, GAS6 and AXL were increased in COVID-19 patients compared to controls. These measures associated with occurrence of clinical events and intensive care unit acquired weakness (ICUAW). cfDNA and GAS6 decreased in time in patients surviving to 30 days post ICU admission. A decrease of 27.2 ng/mL cfDNA during ICU stay associated with patient survival, whereas levels of GAS6 decreasing more than 4.0 ng/mL associated with survival. The presence of H3 in plasma was a common feature of COVID-19 patients, detected in 38% of the patients at ICU admission. NETosis markers cfDNA, H3 and NE correlated well with parameters of tissue damage and neutrophil counts. Furthermore, cfDNA correlated with lowest p/f ratio and a lowering in cfDNA was observed in patients with ventilator-free days.

Figure 1

Detection of plasma markers cfDNA, NE, and Histone H3. Plasma from COVID-19 ICU patients (n = 100), non-COVID-19 ICU patients (n = 11), and healthy control (n = 15) was tested for the presence of cfDNA (A), neutrophil elastase (NE) (B), and extracellular histone H3 (C) at ICU admission. P-values were calculated with the Kruskal–Wallis test with Dunn’s post-hoc test. P-values were considered significant if p < 0.05; * 0.05, ** 0.01, *** 0.001. Statistical analysis of the three groups were p < 0.001 for cfDNA (A); p < 0.001 for NE (B) and p = 0.046 for histone H3 (C).

Figure 2

Detection of plasma markers GAS6 and sAXL. Plasma from COVID-19 IC patients (n = 100), non-COVID-19 IC patients (n = 11), and healthy controls (n = 15) was tested for the presence of GAS6 (A), and sAXL (B) at ICU admission. P-values were calculated with the Kruskal–Wallis test with Dunn’s post-hoc test. P-value were considered significant if p < 0.05; * 0.05, ** 0.01, *** 0.001. Statistical analysis of the three groups were p < 0.001 for Gas6 (A), and p = 0.134 for sAXL (B).

Figure 3

Correlations between pulmonary function and cfDNA in plasma. (A) The lowest p/f ratio measured during stay on the ICU and the cfDNA at ICU admission correlated significantly (r = -0.236; p = 0.021). Correlations were calculated with the Spearman’s rank-order correlation test. Correlations were considered significant if P < 0.05. (B) Change in level of cfDNA in COVID-19 patients on ICU and ventilator free days (VFD). The change in the level of cfDNA in the plasma of COVID-19 patients was calculated by subtracting the average cfDNA level during late (> day 6) from the average level during early days (day 1–5). The patients were divided by VFD = 0 or > 0. P-values were calculated between the VFD groups divided by survival with the Mann–Whitney U test. P-value were considered significant if p < 0.05;*.

Figure 4

Sequential determination of plasma markers cfDNA and Gas6 in COVID-19 patients on ICU and 30-day mortality. Plasma from COVID-19 ICU patients was tested for the presence of cfDNA (A,B), and Gas6 (C,D) during early days (day 1–5), and during late days (> day 6). Survival is based on 30-day mortality. The average plasma marker levels were calculated per group for 33 patients (24 survivors and 8 non-survivors). P-values were calculated between survivors and non-survivors in both early and late days with the Mann–Whitney U test. P-value were considered significant if p < 0.05; * 0.05, ** 0.01, *** 0.001. P-values were calculated between early and late groups divided by survival with the Wilcoxon matched-pairs signed rank test. P-value were considered significant if p < 0.05; ^# 0.05, ^## 0.01, ^### 0.001.

Figure 5

Ideal cutoff value and Kaplan–Meier curves for the prediction of 30-day survival based on sequential levels of (A) cfDNA and (B) Gas6. Receiver operating characteristic (ROC) curve analysis identified the ideal cut-off value with the Youden index of the difference in cfDNA and Gas6 levels in plasma between the early and late time group (n = 33). Kaplan–Meier survival curves for 30-day survival were created based on the identified cutoff value. P-value was calculated with the Mantel-Cox test.