Page 205 - 2021_09-Haematologica-web

Letters to the Editor
Circulating histones play a central role in COVID-19-associated coagulopathy and mortality
COVID-19 has highlighted the lethal consequences of immunothrombosis; i.e., the cross-talk between coagula- tion, inflammation and the innate immune system. Patients with immunothrombosis have significant immune cell death,1 which can release pro-coagulant2 and cytotoxic3 histones. Histones are small, positively- charged proteins that are typically found within the cell nucleus and which bind to negatively-charged DNA. We hypothesize that circulating histones play a central role in critically-ill COVID-19 patients. This translational study demonstrates that admission histone levels are signifi- cantly elevated with increasing severity of COVID-19 infection (Mild, median=2.6 μg/mL [IQR=0.7-7.6], Moderate, 10.5 μg/mL [3.5-27.2], Critical, 20.0 μg/mL [6.2-33.0], Non-survivors, 29.6 μg/mL [11.2-60.0]; P<0.001). Circulating histones associated with severe coagulopathy, inflammation and organ injury markers, including cardiac troponin. Extracellular histone levels on admission are associated with poor outcomes and inde- pendently predict 28-day mortality of hospitalized COVID-19 patients. This is the first report to indicate that circulating histones, released following immune cell death, may play a central pathological role in severe SARS-CoV-2 infection.
COVID-19 was the cause of more than two million deaths worldwide by February 2021,4 resulting from res- piratory and multi-organ failure,5 with evidence of pul- monary thrombosis at post-mortem.6 These patients have extensive immune cell death,1 a strong acute-phase inflammatory response and coagulopathy, as well as car- diac injury.1,5 Cell death can release histones, and extra- cellular histones are cytotoxic, pro-inflammatory7 and pro-coagulant,2 leading to pulmonary thrombosis.8 Extracellular histones also trigger interleukin-6 (IL-6) release to induce an acute phase response, including ele- vation of C-reactive protein (CRP), which, in turn, reduces histone toxicity.9 High levels of circulating his- tones initiate an alternative coagulation pathway during sepsis,2 mediate multiple organ injury3 and correlate with adverse clinical outcomes, including death.10 We there- fore hypothesized that high levels of histones are present in severe SARS-CoV-2 infection, and act as major media- tors of coagulopathy and mortality in COVID-19 disease.
In this study, adult COVID-19 patients (n=113) were recruited at the Royal Liverpool University Hospital from 30th March 2020 to 16th May 2020. Patients were selected using the ISARIC WHO Clinical Characterisation Protocol for Severe Emerging Infections in the UK. Inclusion criteria were: (1) swab positive or high likeli- hood of infection or (2) ≥1 of the following symptoms: fever ≥38°C, new cough, dyspnea or tachypnea and admitted to a healthcare facility.11 Patients were catego- rized into four groups: 1) Mild (minor respiratory symp- toms to exclude shortness of breath OR incidental find- ing, where the patient required admission to hospital for reasons other than COVID-19 (such as for frailty) and was otherwise asymptomatic of COVID-19); 2) Moderate (dyspnea, i.e., patient symptomatic with short- ness of breath OR hypoxia, defined by oxygen satura- tions on pulse oximeter of ≤93% or requiring supplemen- tary oxygen to maintain oxygen saturations ≥96%); 3) Critical disease (respiratory failure requiring the adminis- tration of continuous positive airway pressure (CPAP) to maintain oxygen saturations ≥96% OR invasive ventila- tion in a critical care setting); 4) Non-survivors (patients
who died within 28 days of hospital admission). Circulating histones were quantified in patient plasma on admission (as described previously)8,12 and associa- tions with severity of infection, coagulation, inflammato- ry and organ injury markers were analyzed. Severity of infection was determined by the patient’s most severe clinical state throughout hospital admission, according to the previously described definitions. Cytokines were measured using a Luminex-based bead array, as per man- ufacturer’s instructions (Thermo-Fisher Scientific). Outcome measures included ventilator-support days, length of hospital stay, and 28-day mortality. Ethical approval was provided by the South Central - Oxford C Research Ethics Committee in England (Ref 13/SC/0149), the Scotland A Research Ethics Committee (Ref 20/SS/0028), and the WHO Ethics Review Committee (RPC571 and RPC572, 25 April 2013). Local approval was granted by the North West - Haydock Research
Ethics Committee (REC reference 20/NW/0332).
The Kruskall-Wallis test was used to compare continu- ous variables, presented as median (interquartile range; IQR); the Fisher Exact/χ2 test for comparison of categori- cal variables, presented as counts (percentage). Circulating histone levels were measured by Western Blot, using purified histone as the standard, and analyzed either as continuous variables or categorized based on a previously-determined threshold for cytotoxicity (30 μg/mL).3,7 The Mann-Whitney U test was used to compare categorical histone levels to continuous clinical variables. Correlation analysis was performed using Spearman’s rank. A Receiver Operating Characteristic (ROC) curve analysis and multivariate regression (adjust- ed for age, gender, ethnicity and co-morbidities) assessed admission histone levels in predicting 28-day mortality. Kaplan-Meier survival curve analysis was performed to analyze the probability of mortality over time. Statistical tests were performed on SPSS (IBM, version 25).
A 2-tailed P value of <0.05 was considered significant. The study involved 113 COVID-19 patients (Table 1): median age 65.0 years (IQR=51.0-78.0 years), 65 patients were male (57.5%), 96 of white ethnicity (85.0%). Disease severity was associated with coagulation activa- tion (Table 1), characterized by elevated D-dimer (P=0.017) and prolonged prothrombin time (P=0.005), and a pro-inflammatory phenotype characterized by ele- vated CRP (P<0.001) and IL-6 (P=0.002) on hospital admission, as well as with hypoxia and cardiac injury (Table 1). The median hospital stay was 10 days (IQR,
3-20 days) and 25 patients (22.1%) died within 28 days. Circulating histone levels on admission were signifi- cantly elevated in COVID-19 patients compared to nor- mal controls and were associated with increasing severity of infection (Figure 1A and B; Healthy controls, medi- an=2.9 μg/mL [IQR=1.5-3.3]; Mild, 2.6 μg/mL [0.7-7.6]; Moderate, 10.5 μg/mL [3.5-27.2]; Critical, 20.0 μg/mL [6.2-33.0]; Non-survivors, 29.6 μg/mL [11.2-60.0]; P<0.001). Circulating histone levels strongly correlated with D-dimer levels (R=0.606), indicating the potential involvement of extracellular histones in COVID-19 coag- ulopathy. Positive association with organ injury markers, including bilirubin (R=0.531), creatinine (R=0.501) and cardiac troponin (R=0.486), indicates the possible role of histone-induced cytotoxicity in multiple organ injury. Strong associations with fibrinogen (R=0.632), CRP (R=0.735) and IL-6 (R=0.677) confirmed histone-initiated acute phase response.9 Negative correlation with lym- phocyte count (R=-0.446) suggests that lymphocyte and other immune cell death might be a major source of cir-
culating histones in COVID-19 infection.
haematologica | 2021; 106(9)
2493