Circulating neutrophil extracellular trap (NET)-forming ‘rogue’ neutrophil subset, immunotype [DEspR+CD11b+], mediate multi-organ failure in COVID-19 - an observational study

Background: Cumulative research show association of neutrophils and neutrophil extracellular traps (NETs) with poor outcomes in severe COVID-19. However, to date, no curative intent therapy has been identified to block neutrophil/NETs-mediated progression of multi-organ dysfunction. Because of emerging neutrophil heterogeneity, the study of subsets of circulating neutrophil-extracellular trap (NET)-forming neutrophils [NET+Ns] as mediators of multi-organ failure progression among patients with COVID-19 is critical to identification of therapeutic targets. Methods: We conducted a prospective observational study of circulating levels of CD11b+[NET+N] immunotyped for dual endothelin-1/signal peptide receptor, (DEspR±) expression by quantitative immunofluorescence-cytology and causal mediation analysis. In 36 consented adults hospitalized with mod-severe COVID-19, May to September 2020, we measured acute multi-organ failure via SOFA-scores and respiratory failure via SaO2/FiO2 (SF)-ratio at time points t1 (average 5.5 days from ICU/hospital admission) and t2 (the day before ICU-discharge or death), and ICU-free days at day28 (ICUFD). Circulating absolute neutrophil counts (ANC) and [NET+N] subset-specific counts were measured at t1. Spearman correlation and causal mediation analyses were conducted. Results: Spearman correlation analyses showed correlations of t1-SOFA with t2-SOFA (rho rS=0.80) and ICUFD (rS=−0.76); circulating DEspR+[NET+Ns] with t1-SOFA (rS= 0.71), t2-SOFA (rS =0.62), and ICUFD (rS =−0.63), and ANC with t1-SOFA (rS=0.71), and t2-SOFA (rS=0.61). Causal mediation analysis identified DEspR+[NET+Ns] as mediator of 44.1% [95% CI:16.5,110.6] of the causal path between t1-SOFA (exposure) and t2-SOFA (outcome), with 46.9% [15.8,124.6] eliminated when DEspR+[NET+Ns] were theoretically reduced to zero. Concordantly, DEspR+[NET+Ns] mediated 47.1% [22.0,72.3%] of the t1-SOFA to ICUFD causal path, with 51.1% [22.8,80.4%] eliminated if DEspR+[NET+Ns] were reduced to zero. In patients with t1-SOFA >1, the indirect effect of a hypothetical treatment eliminating DEspR+[NET+Ns] projected a reduction of t2-SOFA by 0.98 [0.29,2.06] points and ICUFD by 3.0 [0.85,7.09] days. In contrast, there was no significant mediation of SF-ratio through DEspR+[NET+Ns], and no significant mediation of SOFA-score through ANC. Conclusions: Despite equivalent correlations, DEspR+[NET+Ns], but not ANC, mediated progression of multi-organ failure in acute COVID-19, and its hypothetical reduction is projected to improve ICUFD. These translational findings warrant further studies of DEspR+[NET+Ns] as potential patient-stratifier and actionable therapeutic target for multi-organ failure in COVID-19.


Introduction
Activated neutrophils release neutrophil extracellular traps (NETs) -a DNA-weblike structure embedded with neutrophil microbicidal/cytotoxic proteases, enzymes, and decondensed histones -to entrap and eliminate pathogens [1,2]. As a robust but non-targeted pathogen-killing defense mechanism, the microbicidal/cytotoxic components in NETs, like a double-edged sword, can also induce bystander "secondary" injury to vascular endothelia and adjacent cells. These injuries erode capillary-tissue barriers causing multi-organ dysfunction progressing to failure, even if the inciting infection is focal or decreasing [3,4]. Cumulative research shows that increased NET-levels are associated with both severity of infection and risk for tissue injury, as seen in the association of increased NETs with COVID-19 severity [5,6] and high mortality [7].
Preclinical studies show causal pathogenic mechanisms for NETs in SARS-CoV-2 virus infection, COVID-19. SARS-CoV-2 virus induces NET-formation in human healthy volunteer neutrophils, and the formed NETs cause injury in human epithelial and endothelial cells [7][8][9], including acute lung injury [10]. Concordantly, increased NET-levels have been implicated in all the clinical pathologies observed in the spectrum of severe COVID-19 multi-organ dysfunction including thromboses [11], parallel to observations in bacterial pneumonia and sepsis-induced models of acute lung injury or multi-organ failure (MOF) [12,13]. The observed common pathogenic roles of excess NETs in secondary tissue injury, systemic micro-thrombosis or microvascular inflammation and occlusion, suggest reduction of NETs as a potential therapeutic approach to MOF, which to date remains without curative-intent therapy.
However, since NETs provide multiple key defense mechanisms against bacterial infections [2], sepsis [14] and viral infections [15], therapeutic approaches to block NETsmediated secondary "bystander" tissue injury must target dysregulated NET-formation, but spare homeostatic regulated NET-formation. Given neutrophil heterogeneity and multiple mechanisms of NET-formation [16,17], identification of dysregulated "rogue" NET-forming neutrophil subsets/subtypes that escape normal NET-clearance, accumulate, and contribute to multi-organ failure could be key to much-needed targeted therapies for severe  Our recent studies have identified DEspR+CD11b+ neutrophils (DEspR+[Ns] from hereon), as a dysregulated "rogue" neutrophil-subset capable of NET-formation while in the circulation, extended survival, low-clearance, enhanced adhesion, and association with severity measures and mortality in COVID-19 acute respiratory distress syndrome (ARDS) [18][19][20].
Here, we test whether the putative "rogue" CD11b+DEspR+ NET-forming neutrophil-subset, , mediates the worsening of multi-organ dysfunction (as measured by the Sequential Organ Failure Assessment [SOFA]-score) [21] and poor clinical outcomes (as measured by intensive care unit free days [ICUFD]) [22] in severe COVID-19 using causal mediation analysis. Identification of a rogue [NET+N] subset that mediates progression of multiorgan dysfunction in severe COVID-19 patient samples has the potential to identify a muchneeded therapeutic target and/or biomarker. The combinatorial use of direct analysis of patient neutrophils and causal inference mediation statistics has the potential to validate a translational approach to overcoming low translatability of animal models in ARDS-multi-organ failure regardless of etiology.

Methods
Study design and Participants. Procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional or regional) and with the Helsinki Declaration of 1975, as most recently amended (https://www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-humansubjects/).
A combined 2-site analysis of NET-forming neutrophil subsets from two independent prospective observational study cohorts previously characterized for 'rogue' neutrophils and NETosis [18,19]. At Boston Medical Center (BMC), the protocol study number is H-36744, study title: Humanized anti-DEspR antibody therapy for Acute Lung Injury (ALI/Acute Respiratory Distress Syndrome), and approval by Boston University's Institutional Review Board on 12-01-2019. At Maine Medical Center (MMC), the protocol study number is 1598969-16, with study title "IT-19 Identification of molecular treatment targets for COVID-19", and approval by Maine Medical Center's Institutional Review Board on 5/8/2020. Clinical data and blood sample collections followed IRB approved protocols, (Supplemental Methods). Informed consent was obtained from the patient when able, or when unable, from the patient's legally authorized representative (LAR). LAR-informed consents were obtained by phone using an IRBapproved informed consent-by-phone at BMC, or electronically at MMC, to minimize viral exposure. Clinical parameters of severity were obtained: non-neurologic SOFA score as a measure of multi-organ dysfunction and SaO2/FiO2 (SF)-ratio as a measure of respiratory distress/failure, and ICU-free days at day 28, with competing risk of death -1 (ICUFD) (22) as a summation outcome measure. Clinical measures were taken from two time points: timepoint-1 (t1) after informed consent after COVID19 diagnosis verification upon admission to the hospital or ICU, average 5.5 days, and timepoint-2 (t2): the day before ICU-death or ICU/hospital discharge. CBC-differential and blood samples for immunofluorescence cytology were obtained at timepoint-t1.

Semi-automated Quantitation of [NET+Ns] subsets. Third party Nikon Imaging
Laboratory (Cambridge MA) performed quantitative imaging analysis blinded to patient information for automated unbiased detection and quantitation of NET-forming DNA-extruding neutrophils by measuring the circularity index using a standard shape analysis formula 4pi (area/perimeter 2 ) validated earlier [19]. Quantitative analysis of subsets of NET-forming neutrophils with molecular markers for CD11b± and DEspR± and DNA was performed via automated determination of fluorescence intensities of NET-forming neutrophils identified by circularity index < 0. Causal Mediation analysis (CMA) was performed using R (version 4.2.1) and package regmedint [25]. Causal mediation analysis seeks to disentangle relationships between three or more variables [26][27][28] where some or all of the effect of an exposure (A) on an outcome (Y) is counts due to a moderate change in the exposure variable (i.e., from its first quartile value to its third quartile value). The pure natural direct effect (the direct effect of exposure on outcome if the mediator [NET+Ns] is set at the value it would naturally take when exposure is at its reference value), the total natural direct effect (the direct effect of exposure on outcome accounting for exposure and mediator interaction), and the total effect (total effect of exposure on outcome through direct and indirect pathways) were also reported. As comparator, we also tested absolute neutrophil counts (ANC) as a mediator of t1-SOFA to t2-SOFA and t1-SOFA to ICUFD. We used bootstrapping with 10,000 replicates to calculate 95% confidence intervals.

Analysis of effects of a hypothetical treatment reducing DEspR+[NET+Ns] as mediator.
We estimated the indirect effect on t2-SOFA, ICUFD, and t2-SF ratio mediated by DEspR+[NET+Ns] of a hypothetical treatment that would eliminate DEspR+[NET+Ns] as described by Lok and Bosch [29], which showed that a treatment effect on the mediator can be combined with off-treatment mediator and outcome data to estimate a causal indirect effect (but not a total or direct effect). To estimate the indirect effect on t2-SOFA and ICUFD mediated by and t2-SOFA/ICUFD, and also included t1-SOFA ≥ 4 as a second pre-treatment causal variable.
For the t2-SF ratio, we included the t1-SF ratio as a pre-treatment common cause. For these analyses we modified existing SAS macros (29), for a continuous outcome modeled with linear regression similar to regmedint [25].

Causal Mediation Analysis of circulating DEspR+[NET+N] levels
Based on preclinical causal relationships of SARS CoV2, NET-formation, and injury to endothelial-tissue barrier in different vital organs contributing to multi-organ failure and supported by corresponding correlation analyses depicting linear relationships ( Table 2, Supplemental Fig. S1), we tested causal path hypotheses for progression of multi-organ failure ( Fig. 3-A) and progression of respiratory failure (Fig. 3-B). As shown in  (Table 3), indicative of a decrease in progression of multi-organ failure.
In contrast, mediation analysis for progression to acute respiratory failure (Fig. 3) showed no significant mediation by DEspR+[NET+Ns] between t1-SF and t2-SF, and between t1-SF and ICUFD (Table 3). Additionally, despite strong correlations with t2-SOFA and ICUFD (Table 2), there was no significant evidence of mediation by ANC on both causal path hypotheses for progression of multi-organ failure (Fig. 3-A): from t1-SOFA to t2-SOFA and from t1-SOFA to ICUFD (Table 3).

Discussion
Premised upon preclinical studies detecting a causal role of SARS CoV2 on NETformation and NET-forming neutrophils on capillary-tissue barriers, [8,9] [20]. We note that while causal mediation analysis has been previously used to evaluate inflammatory mediators of the effect of obesity on risk for mortality in COVID-19 [31], and soluble RAGE receptor levels and angiopoietin-2-levels in sepsis-related ARDS [32,33], here we report causal mediation analysis of NET-forming neutrophil subsets.
The differential mediation effect profiles for DEspR+[NET+Ns] compared with ANC, despite similarly strong correlations with multi-organ failure outcomes, highlight the emerging role of NET+Ns as a central mechanism for neutrophil-mediated secondary tissue injury and/or immunothrombosis in multi-organ failure in COVID-19 [7,[34][35][36]. Likewise, the differential mediation of worse SOFA score and ICUFD -but not of worse SF-ratio -by DEspR+[NET+Ns] -indicates specificity of causal effect estimates of mediation. This differential mediation supports the hypothesis that the low SF-ratio in COVID-19 is caused by the direct damage of respiratory epithelia infected with the SARS CoV2 virus [37], rather than by neutrophil-mediated tissue injury of indirect ARDS [38]. Taken together, concordance of our findings supports the pathogenic role of circulating NET+Ns in direct endothelial injury and microcirculation compromise in the progression of secondary multi-organ failure in COVID-19.
In the analysis of a hypothetical treatment that eliminates DEspR+[NET+Ns] among patients with a SOFA score of 2 or more, elimination of DEspR+[NET+Ns] was associated with an indirect effect of a 1-point decrease in subsequent SOFA score, which for patients with COVID-19 ARDS, would translate approximately to a 15% absolute risk reduction (ARR) in death [39]. Similarly, elimination of DEspR+[NET+Ns] was associated with a 3-day increase in ICUFD. These results suggest that a novel treatment that eliminates DEspR+[NET+Ns] could potentially reduce mortality in severe COVID-19 with an effect estimate at least as strong as that of corticosteroids [40,41].
Notably, preclinical studies support the feasibility of this therapeutic hypothesis as the anti-DEspR antibody induces apoptosis in DEspR+ neutrophils observed on live cell imaging of macaque neutrophils and promotes neutrophil apoptosis without worsening elevated myeloperoxidase and complement activation levels, as observed in an ex vivo experimental system testing ARDS patient whole blood samples [19]. Intuitively, the induction of apoptosis in circulating "rogue" [DEspR+CD11b+] neutrophils will pre-empt DEspR+  [29].

Limitations of study:
Our small cohort size may be underpowered to detect weaker associations. This observational study was done during the first phase of COVID-19 without vaccination or antiviral therapies available; thus, how these therapies may modify the relationships between organ dysfunction and NETs is unclear. Since NET levels vary with time and t1 was not the same for all patients due to informed consent issues, the study design may have introduced t1 noise/variation that decreases power.

Conclusions and Clinical Implications:
In this prospective pilot study, Causal Mediation Analysis detected DEspR+ [NET+N] subset as a mediator of progression of multi-organ-dysfunction in COVID-19 and the hypothetical reduction of DEspR+[NET+N] subset support the therapeutic hypothesis that prevention or reduction of DEspR+[NET+Ns] has the potential to reduce progression to multiorgan failure in severe acute COVID-19. Altogether, data provide translational milestones in support of further studies to advance DEspR+[NET+Ns] as a much-needed potential biomarker for patient stratification and therapeutic target for multi-organ failure in severe acute COVID-19.

Ethics approval and consent to participate
Procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional or regional) and with the Helsinki Declaration of 1975, as most recently amended (https://www.wma.net/policies-post/wma-declaration-of-helsinkiethical-principles-for-medical-research-involving-human-subjects/). At Boston Medical Center (BMC), the protocol study number is H-36744, study title: Humanized anti-DEspR antibody therapy for Acute Lung Injury (ALI/Acute Respiratory Distress Syndrome), and approval by Boston University's Institutional Review Board on 12-01-2019. At Maine Medical Center (MMC), the protocol study number is 1598969-16, with study title "IT-19 Identification of molecular treatment targets for COVID-19", and approval by Maine Medical Center's Institutional Review Board on 5/8/2020. Clinical data and blood sample collections followed IRB approved protocols.

Consent for publication.
Not applicable.

Availability of data and materials
Data are available from corresponding author upon reasonable request.

Competing interests
Boston University holds awarded patents on DEspR as novel therapeutic approach for cancer, stroke, COVID-19, NETosis and activated neutrophils. VLMH and NRO are co-inventors in patents filed by Boston University.