key: cord-0950563-u4958nsb
authors: Gavriilidis, E.; Antoniadou, C.; Chrysanthopoulou, A.; Ntinopoulou, M.; Smyrlis, A.; Fotiadou, I.; Zioga, N.; Kogias, D.; Natsi, A.-M.; Pelekoudas, C.; Satiridou, E.; Bakola, S.-A.; Papagoras, C.; Mitroulis, I.; Peichamperis, P.; Mikroulis, D.; Papadopoulos, V.; Skendros, P.; Ritis, K.
title: Combined administration of inhaled DNase, baricitinib and tocilizumab as rescue treatment in COVID-19 patients with severe respiratory failure
date: 2022-03-15
journal: nan
DOI: 10.1101/2022.03.14.22270915
sha: fe2d997c443bbf99410d9a6232ddd453fcc9570f
doc_id: 950563
cord_uid: u4958nsb

COVID-19-related severe respiratory failure (SRF) leads to mechanical ventilation increasing the in-hospital mortality substantially. Abundancy of lung fibroblasts (LFs) in injured lung tissue has been associated with the progression of respiratory failure in COVID-19. Aiming to reduce mortality in patients with SRF (PaO2/FiO2<100 mmHg) and considering the multi-mechanistic nature of severe COVID-19 pathogenesis, we applied a combined rescue treatment (COMBI) on top of standard-of-care (SOC: dexamethasone and heparin) comprised inhaled DNase to dissolve thrombogenic neutrophil extracellular traps, plus agents against cytokine-mediated hyperinflammation, such as anti-IL-6 receptor tocilizumab and selective JAK1/2 inhibitor baricitinib. COMBI (n=22) was compared with SOC (n= 26), and with two previously and consecutively used therapeutic approaches, including either IL-1 receptor antagonist anakinra (ANA, n=19), or tocilizumab (TOCI, n=11), on top of SOC. In parallel, evaluation of immunothrombosis was assessed in vitro in human LFs, treated with the applied therapeutic agents upon stimulation with COVID-19 plasma. COMBI was associated with lower in-hospital mortality (p=0.014) and intubation rate (p=0.013), shorter duration of hospitalization (p=0.019), and prolonged overall survival after a median follow-up of 110+/-4 days (p=0.003). In vitro, COVID-19 plasma markedly induced tissue factor/thrombin pathway in LFs, while this effect was inhibited by the immunomodulatory agents of COMBI providing a mechanistic explanation for the clinical observations. These results suggest the design of randomized trials using combined immunomodulatory therapies in COVID-19-associated SRF targeting multiple interconnected pathways of immunothrombosis.

In patients with COVID-19, severe hypoxemic respiratory failure (SRF) leading to invasive mechanical ventilation (IMV), raises the mortality rate substantially [1] [2] [3] . Thus, the management of patients with SRF to avoid intubation and intensive care admission is a challenging and crucial issue [2] [3] [4] . Numerous clinical trials were conducted or are underway in the efforts to improve the current mainstay of treatment for patients with severe COVID- 19 , which consists of dexamethasone, heparin, and supportive care [5, 6] .

Deterioration of COVID-19 respiratory function seems to be associated with progressed dysregulation of immune responses [7] . Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infects predominantly alveolar epithelial cells, triggering aberrant inflammatory responses leading to acute respiratory distress syndrome (ARDS) [8, 9] .

Neutrophil extracellular traps (NETs) are related with the activity of various immunological mediators and the subsequent inflammatory milieu [10] . Moreover, advanced respiratory disease in COVID-19 is characterized by macrophage extravasation, while an abundance of mesenchymal cells and fibroblasts is observed and linked with the progression of respiratory failure [8, 9] .

Mechanistically, NET-mediated immunothrombosis, leads to activation of tissue factor (TF) axis and thrombin generation. Proteases of this axis fuel a vicious cycle, in which a harmful crosstalk between thrombosis, inflammation and fibrosis is developed [8, 11] . Moreover, complement activation, triggering of the NF-κB and JAK/STAT pathways, have been described as partners in COVID-19 hyperinflammation [11] [12] [13] . Based on these mechanisms and the current ongoing randomized control trials (RCTs), immunomodulatory treatments including anakinra, a recombinant IL-1 receptor antagonist [14] , tocilizumab, an anti-IL-6 receptor antagonist [15] , baricitinib, a selective JAK-1/JAK-2 inhibitor [16] and nebulized 4 recombinant human DNase [17, 18] as an agent which dismantles the generated NETs, have been administered separately in patients with COVID-19-related SRF.

This study describes, as rescue treatment, a therapeutic protocol based on the multimechanistic nature of severe COVID-19. Standard-of-care (SOC) in conjunction with a combined therapeutic strategy, comprising inhaled DNase plus baricitinib and tocilizumab, was found to reduce the mortality in a group of patients with severe respiratory failure.

Translating the clinical findings, we observed diminished expression of TF/thrombin axis in primary lung fibroblasts treated with the applied therapeutic agents upon stimulation with COVID-19 environment.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022 were consecutively admitted from October 25, 2020 to August 18, 2021. Among them, we studied 78 patients who progressed to severe respiratory failure (SRF) defined as a ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO2/FiO2) <100 mmHg [2] . They received SOC treatment at the time of admission or combination of SOC with known immunomodulating factors, in a consecutive chronologically pattern, as compassionate/rescue treatment (Figure 1 ). 

The selection of the different rescue treatments that have been used throughout the study period, was initially decided, and periodically revised, by considering any new published or ongoing RCT, as well as emerging research findings of our group and others on the role of multiple pathways in the inflammatory milieu of severe COVID-19, including the critical role of NETs, the triggering of NFκB pathway and the induction of IL-6 and Janus kinases (JAKs). In total, 4 SRF groups were compared and analyzed, including the group of patients treated with SOC only and 3 consecutive chronologically additional groups in which apart from SOC treatment, different rescue therapeutic immunomodulatory regimens have been added (Figure 1 ).

SOC group comprised patients who received only standard of care in accordance to our department protocol including dexamethasone 6-8 mg once daily, low molecular weight heparin at therapeutic doses, antibiotic prophylaxis and supportive care.

TOCI group included patients who received SOC plus IL-6 receptor inhibitor tocilizumab.

Tocilizumab was administered as a single intravenous (iv) dose of 8 mg/kg actual body weight up to 800 mg.

ANA group included patients who received SOC plus IL-1 receptor antagonist anakinra.

Anakinra was administered iv 200 mg/twice daily for 3-6 days, then 100 mg/twice daily, for up to 10 days in total.

COMBI group included patients who were treated with SOC and the combination following regimens: a) tocilizumab (as described above), b) selective JAK-1/JAK-2 inhibitor baricitinib, 4 mg per os once daily, for up to 14 days (2 mg if GFR was 30 to <60 ml/min/1.73 m 2 ) and c) tο dismantle the accumulated NETs in lung, nebulized dornase alfa, a recombinant human DNase I (inhaled DNase), 2,500 U/twice daily, for up to 14 days. Inhaled DNase was . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint administered simultaneously with inhaled budesonide (800 µg/twice daily) and bronchodilators (salbutamol or/and ipratropium at conventional doses).

During the course of the study, the same general care protocols were used by the nursing personnel. Baseline demographical and clinical characteristics of the patients are demonstrated in Table 1 .

Primary endpoint was the reduction of the in-hospital mortality rate, whereas secondary endpoints concerned intubation/IMV rate, days of hospitalization and overall survival as derived from the last follow-up visit, either in-office or remote.

The study was aligned with the Helsinki Declaration. Subjects provided written informed consent before study participation. Patients' records were anonymized and deidentified prior to analysis and so confidentiality and anonymity were assured. The study protocol design was approved by the local scientific and ethics committees and institutional review board of the University Hospital of Alexandroupolis (Ref. No. 87/08-04-2020).

Chi square test along with adjusted residuals was used to compare outcomes among the four treatment groups. Similarly, chi square and ANOVA were used to compare binary and discrete/continuous variables, respectively, that are considered to be potential confounders.

To consolidate the independent correlation of each treatment group with outcome, a Generalized Linear Model using outcome as dependent variable, treatment groups as independent variables and potential confounders as factors was further utilized. For that purpose, all scale variables were turned into binary ones with the use of Optimal Scaling along with ridge regression, random initial configuration and bootstrapping. Secondary outcomes were evaluated with ANOVA; ad-hoc analysis was performed using Tuckey's HSD . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint test. The repeated measures General Linear Model was used for analysis of within-subjects and between-subjects variance of the same variable measured several times on each patient.

Kaplan-Meier curves were used to depict survival data; comparisons were performed by the Logrank test. A Cox proportional-hazards regression model was introduced to examine simultaneously the effects of multiple covariates on overall survival. Median follow-up was approached by the reverse Kaplan-Meier estimator.

Comparisons between two groups were performed using Student's t test (2-tailed). For comparisons among more than two groups, Kruskal-Wallis test, followed by Dunn's test for multiple comparisons, was performed.

For both clinical and experimental results, the level of statistical significance was set at 0.05.

Bonferroni correction was applied in case of multiple testing. Data are presented as mean ± standard deviation (SD). Statistical analysis was performed using SPSS 26.0 software.

Detailed information for all methods can be found in the in the Supplementary Materials section.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint 3. Results

Analysis of baseline characteristics revealed that all groups of SRF patients were comparable at enrollment ( Table 1 ).

In the SOC group, 9 deaths were recorded among 26 patients (34.6%). 10 out of 26 patients were intubated (38.5%); 8 of them did not survive. The mean duration of hospitalization was 19.4 days (SD: ±7.2). In the ANA group, 6 deaths and 6 intubations were recorded among 19 hospitalized patients (31.6%). Five intubated patients did not survive. The mean duration of hospitalization was 23.9 days (SD: ±10.8). In the TOCI group, 7 deaths were recorded out of 11 patients, all who had been intubated (63.6%). The mean duration of hospitalization was 19.4 (SD: ±7.8). In the COMBI group, there were 2 deaths among 22 patients (9.1%), both in intubated patients. There was a reduction in the mean duration of hospitalization, which was 15.6 days (SD: ±5.3).

Taken together, COMBI treatment was associated with significant lower in-hospital mortality (p=0.014), as well as intubation rate (p=0.013) and shorter duration of hospitalization (p=0.019) compared to other therapeutic groups of patients with COVID-related SRF (Figure   2A ,B). Moreover, COMBI treatment was correlated with prolonged survival (p=0.003) after a median follow-up of 110±4 days; loss-of-follow-up was 1.3% (1 patient at SOC group) ( Figure 2C) .

A Generalized Linear Model was used to assess the effect of potential confounders in mortality rate; similarly, a Cox regression proportional hazards model, incorporating the same potential confounders as covariates, was introduced to approach survival probability in a stochastic manner. Taken together, these data further supported the initial findings (Supplementary Table 1, Supplementary Figure 1) .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint

Overall immunomodulatory treatments were well tolerated. All the recorded severe adverse events were related to severe COVID-19 disease (Supplementary Table 2 ). There were no serious infections associated with the immunomodulatory therapies.

There is sufficient evidence showing that inflammatory and hematological markers such as the absolute neutrophil and lymphocyte count (ANC, ALC), LDH, CRP and D-dimers, are associated with the progression and severity of COVID-19 [20] [21] [22] . As a result, we recorded these markers on the first day of progression to SRF, as well as on day 7 after the initiation of each treatment. Comparison between groups, showed statistically significant results in favor of the COMBI group, in the increase of ALC count (p=0.021) and the reduction of CRP levels (p=0.002) on day 7 (Figure 3A,B) . There was no statistically significant difference between groups, regarding ANC, LDH and D-dimers values.

Since immunothrombosis is crucially involved in pathology of COVID-19 ARDS [23] , mesenchymal cell/fibroblast accumulation in lung is linked with the progression of COVID-19 severe respiratory failure [8] and fibroblasts under certain inflammatory conditions express TF [24, 25] , we examined whether COVID-19 environment could activate TF/thrombin pathway in LFs. We observed that plasma samples from treatment-naïve COVID-19 patients markedly induced TF expression in LFs, compared to untreated cells, as indicated by TF realtime quantitative PCR (qPCR), in-cell ELISA and immunofluorescence microscopy ( Figure   4A ,B and D). TF released by plasma-stimulated LFs was bioactive, as assessed by TF activity . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. quantitative assay ( Figure 4C) . Together, our findings suggest that COVID-19 inflammatory microenvironment is a potent activator of the thrombotic potential of LFs.

According to previous findings, and considering that the proteases of TF/thrombin axis have a multivalent function leading to thrombosis, amplification of inflammation, cell proliferation, and fibrosis [25] [26] [27] , we prompted to study the in vitro effect of the applied therapeutic regimens.

SARS-CoV-2 infection seems to activate the NF-kB signaling pathway, which may subsequently induce the secretion of multiple inflammatory cytokines, including interleukin (IL)-1, interleukin (IL)-6 and tumor necrosis factor-α (TNF-α) [28] . Since both IL-1 and IL-6 are elevated in patients with COVID-19 [14, 15, 29] , LFs were pretreated with either a recombinant human IL-1 receptor antagonist (anakinra) or an anti-IL-6 receptor monoclonal antibody (tocilizumab), in an attempt to disrupt the autoinflammatory loops driven by these cytokines. However, TF expression ( Figure 5A On the other hand, pre-incubation of cells with a selective JAK1/JAK2 inhibitor with an immunosuppressive role (baricitinib) [16] , or pre-treatment of plasma samples with DNase I, an enzyme able to dismantle NETs [17, 18] , markedly reduced TF expression (Figure 5A- In view of these data and considering that COVID-19 is mediated by plenty of plausible mechanisms, cells were simultaneously blocked against IL-6, JAK1/JAK-2 and NET signaling, leading to TF suppression in plasma-stimulated LFs (Figure 5A-D) .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint

COVID-19 patients with SRF, are in immediate need of therapeutic solutions to avoid mechanical ventilation or death. This study focused on that difficult group of patients, establishing hard inclusion criteria and endpoints. Considering the involvement of multiple mechanisms in COVID-19 pathology, it is a rational approach to apply combined therapeutic strategies aiming at reducing mortality. In this context, keeping dexamethasone and heparin as SOC treatment, we observed benefit regarding the final outcome, namely in-hospital mortality, in SRF patients by adding inhaled DNase combined with agents against IL-6 and JAK1/2 signaling. Moreover, treatment of primary lung fibroblasts, main players during the progression of respiratory failure [8, 30] , with the administrated immunomodulatory agents upon stimulation with COVID-19 environment, disrupted TF/thrombin expression, providing potential mechanistic insights regarding their therapeutic effect.

After the infection of respiratory epithelial cells by SARS-CoV-2, the sequence of events regarding the cross-talk of the involved inflammatory mediators, remains an obscure scientific field. A large amount of evidence has suggested the triggering of various amplification loops in a vicious cycle pattern. It is highlighted that viral proteins trigger the activation of NFκB pathway [31] and complement [11, 32] . The activation of NFκB, induces the secretion of multiple chemokines and cytokines such as IL-1, IL-6, IL-8 and TNFα. Cytokine-mediated amplification loops enhance this pathway, leading to a true "cytokine storm" and influx of macrophages and neutrophils damaging the lung tissue [28] . Additionally, mediators of the inflammatory environment of COVID 19, such as, complement anaphylatoxins, plateletderived factors, thrombin, IL-8 and G-CSF, trigger the NET-driven thromboinflammation [23, [32] [33] [34] . Into the NET structures, proteases and cytokines are preserved in their active form, further amplifying thrombotic tendency, inflammation, and fibrosis [11, 35, 36] .

Moreover, IL-6 triggers JAK/STAT signaling, driving the transcription of an extensive range . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint of acute phase proteins such as CRP. Furthermore, IL-6 may activate and promote differentiation or proliferation of several non-immune cells, including fibroblasts [37, 38] .

The application of combined therapeutic regimens against COVID-19-related SRF is an emerging challenge against this complex and heterogeneous mechanistic basis of COVID-19 pathophysiology.

Heparin, as a part of SOC regimen, by potentiating antithrombin III (ATIII) activity, inhibits the hypercoagulable condition and the generation of fibrin microthrombi in the lung [39, 40] .

Of note, in COVID-19 low circulating levels of ATIII have been described, making the potency of heparin in these cases defective [41] [42] [43] [44] . However, heparin may also exert its beneficial effects against immunothrombosis by dismantling thrombogenic NETs [45] . In the same context, the other partner of SOC, dexamethasone, the first agent proved to reduce mortality in a RCT [5] , has anti-inflammatory/anti-cytokine effects contributing to the control of NFκB-induced hyperinflammation [46] . Recently it has been suggested that dexamethasone, by modulating COVID-19 immature neutrophils, favors the immune response [47] . In this study, we initially added on SOC, immunomodulating monotherapies using anakinra or tocilizumab consecutively.

Anakinra and tocilizumab have been tested in several RCTs that enrolled patients with COVID-19-related SRF. CORIMUNO-ANA-1, compared the use of anakinra to usual care in 116 hospitalized patients who were hypoxemic but did not require high-flow oxygen or ventilation, was stopped early for futility [48] . REMAP-CAP found that anakinra was not effective in reducing the combined endpoint of in-hospital mortality and days of organ support in ICU patients with COVID-19 [49] . Recent results of the RECOVERY and REMAP-CAP trials provided evidence that tocilizumab, when administered with corticosteroids, offers a modest mortality benefit in certain patients (including ICU patients)

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint with COVID-19 who are severely ill, who are rapidly deteriorating and have increasing oxygen needs, and who have a significant inflammatory response [49, 50] .

In this study that recruited the most severely affected patients (PaO2/FiO2 <100 mmHg), the addition of anakinra or tocilizumab on top of SOC did not add significant benefit regarding mortality, need for IMV and time to discharge. As the mortality rates observed remained high, we considered that the mechanisms involved in the inflammatory environment of COVID-19 are probably interdependent and create a vicious cycle of positive feedback.

In the next, (fourth), consecutive therapeutic protocol, we added on SOC treatment a combined immunomodulation scheme, which comprised inhaled DNase to dismantle NETs and locally suppress the activity of their proteins, tocilizumab to inhibit the signalling of preexisting IL-6 load, interrupting an initial link between JAK/STAT and NFκB, and baricitinib to further prevent the JAK/STAT dependent transcription [15, 16, 28, 35, 45, 51, 52] . Using this protocol (COMBI), the in-hospital mortality rate of this group of COVID-19 patients with SRF was reduced to 9.1%. In line with our clinical observations, COV-BARRIER RCT recently reported an additional survival benefit of baricitinib when added to corticosteroids in hospitalized, non-IMV, patients with COVID-19 pneumonia. The effect was most pronounced in the subgroup of patients receiving high-flow oxygen or non-invasive ventilation at baseline [53] . Additionally, it has been suggested that baricitinib has antiviral activity through interference with SARS-CoV-2 endocytosis [54] .

Currently, clinical data regarding the inhaled DNase in COVID-19 are very limited, coming only from two small, non-randomized, case-control trials enrolled ICU patients with ARDS [17, 18] . Treatment with nebulized DNase in most of the patients was associated with improved oxygenation and outcome, especially when used earlier in the disease course [17, 18] . Decreased NETs remnants in bronchoalveolar lavage fluid was also described [18] .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint Patients received also corticosteroids, and in some cases convalescent plasma and anticoagulation [17, 18] .

Administration of COMBI led to a rapid reduction of CRP and restoration of lymphopenia, both systemic biomarkers of disease severity and progression in COVID-19 [20] [21] [22] . This is in agreement with the favorable disease outcome of these patients. Furthermore, recovery of lymphocyte count probably implies a reversal of the impaired adaptive cellular immune response described in severe COVID-19 patients [55] .

Reduced mortality of SRF patients in COMBI group, prompted us to investigate the in vitro inhibitory effect of applied immunomodulatory agents, namely DNase, tocilizumab, anakinra, and baricitinib in COVID-19 culture conditions. Lung fibroblasts have been found to significantly accumulate and proliferate, repopulating the damaged alveolar wall as the disease progresses, however, their functional role in COVID-19 is still elusive [9, 30, 56] .

Recently, Rendeiro et al described the spatial landscape of COVID-19 lung pathology in COVID-19-decedents by correlating clinical and pathological variables with high-parameter imaging mass cytometry data. Fibroblasts were closely associated with alveolar type-2 cells, large thrombi, and interventions such as intubation and treatment [8] . Moreover, S and N proteins of SARS-CoV-2 are able to induce in vitro the expression of TF on human fibroblasts [57] . Taken together, it appears that fibroblasts emerge as major players of lung tissue thromboinflammation in progressive COVID-19.

We observed an inhibitory effect on TF/thrombin axis expression and activity by primary lung fibroblasts after their treatment with the above immunomodulatory agents, used separately or combined as in the COMBI therapeutic protocol. Dismantling of NETs and inactivation of the mediators they contain, as well as JAK inhibition, seem to have a central role in countering NET driven thromboinflammation. The observed clinical outcome in all groups and the improved regulation of TF activity in supernatants obtained after treatment of . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint cells with DNase, or/and IL-6 and JAK1/2 inhibition, indicate the involvement of various inflammatory triggers and pathways in TF/thrombin axis activation. This suggests the potential value of novel strategies based on combined immunomodulation. Several studies provide consistent evidence that TF/thrombin pathway, apart from thrombosis, is involved in inflammation and fibrotic process [27, 58] . Mechanistically, fibroblasts-derived extracellular vesicles/exosomes or microparticles could transfer active TF to the lung tissue environment [59] [60] [61] . According to recent studies [11, 13, 32] , combined targeting against other involved mediators, specifically in early stages, such as complement anaphylatoxins, may further diminish the mortality rate.

In conclusion, in this study, including non-ICU patients with SRF secondary to COVID-19 pneumonia, we found that combined compassionate therapy using inhaled DNase, tocilizumab, and baricitinib on top of SOC resulted in lower mortality and intubation rate, as well as shorter hospitalization time, compared to the use of SOC alone or along with monotherapies. Inhibition of TF/thrombin axis in lung fibroblast offers a plausible mechanistic explanation for these clinical observations. These results, although in a singlecenter non-RCT, encourage the design of RCTs using combined immunomodulatory therapies, which can target in a multilevel manner interconnected mechanisms of COVID-19 pathogenesis, in patients with the highest disease severity, before they cross the ICU road.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint Figure 1. Consecutive groups of therapies and number (n) of patients enrolled during the study period; the symbol + denotes in-hospital deaths.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint Figure 4 . COVID-19 plasma triggers lung fibroblasts to produce tissue factor (TF) in vitro. Lung fibroblasts (LFs) were stimulated with 2% plasma, derived either from healthy subjects (healthy plasma) or COVID-19 patients (COV plasma). Tissue factor (TF) expression in LFs as assessed by (A) qPCR and (B) In-Cell ELISA (Cytoblot). All conditions were compared to untreated cells. (C) TF activity was evaluated in culture supernatants of LFs as well, in COVID-19 plasma samples as control, diluted directly with the culture medium in a final dilution that used in experimental conditions. In (A)-(C), n=15, bars represent mean±SD, (n.s.: not significant); in case of (C), alpha was set to 0.0125 after Bonferroni correction. (D) Confocal fluorescence microscopy showing TF in LFs (green: TF, blue: DAPI). A representative example of three independent experiments is shown.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint Figure 5 . Agents of combined treatment result in reduction of tissue factor (TF) expression and activity in cultures of lung fibroblasts. Relative fold expression of (A) tissue factor (TF) mRNA and (B) In-Cell TF ELISA (Cytoblot) in lung fibroblasts (LFs) treated with 2% COVID-19-derived plasma (COV plasma) and inhibited with a recombinant IL-1 receptor antagonist (anakinra), an anti-IL-6 receptor monoclonal antibody (tocilizumab), a selective JAK1/JAK2 inhibitor (baricitinib), DNase I or combination of therapeutic agents (tocilizumab, baricitinib and DNase I). (C) TF activity in cell supernatants in conditions as previously described. In (A)-(C), the effect of therapeutic agents was compared to COV plasma condition, n=6, bars represent mean±SD, statistical significance was set at p<0.05 (n.s.: not significant). (D) Confocal fluorescence microscopy showing TF staining in stimulation and inhibition studies of LFs (green: TF, blue: DAPI). A representative example of three independent experiments is shown.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 15, 2022. ;  https://doi.org/10.1101/2022.03.14.22270915 doi: medRxiv preprint

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention

Severe Covid-19

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

and the Northwell COVID-19 Research Consortium, Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the

Dexamethasone in Hospitalized Patients with Covid-19

Surviving Covid-19 with Heparin?

COVID-19 Immunobiology: Lessons Learned, New Questions Arise

The spatial landscape of lung pathology during COVID-19 progression

Immunopathology of Hyperinflammation in COVID-19

Complement and tissue factor-enriched neutrophil extracellular traps are key drivers in COVID-19 immunothrombosis

The Many Faces of JAKs and STATs Within the COVID-19 Storm

Complement activation in patients with COVID-19: A novel therapeutic target

Anakinra for severe forms of COVID-19: a cohort study

Baricitinib, a drug with potential effect to prevent SARS-COV-2 from entering target cells and control cytokine storm induced by COVID-19

The Use of Dornase Alfa in the Management of COVID-19-Associated Adult Respiratory Distress Syndrome, Critical Care Research and Practice

Non-Randomized Trial of Dornase Alfa for Acute Respiratory Distress Syndrome Secondary to Covid-19

Introducing the Palliative Performance Scale to clinicians: the Grampian experience

Prognostic value of lymphocyte count in severe COVID-19 patients with corticosteroid treatment

The Activin/Follistatin Axis Is Severely Deregulated in COVID-19 and Independently Associated With In-Hospital Mortality

Biomarkers and outcomes of COVID-19 hospitalisations: systematic review and metaanalysis

Endothelial dysfunction and immunothrombosis as key pathogenic mechanisms in COVID-19

Elevated Circulatory Levels of Microparticles Are Associated to Lung Fibrosis and Vasculopathy During Systemic Sclerosis

Tissue factor-thrombin signaling enhances the fibrotic activity of myofibroblasts in systemic sclerosis through up-regulation of endothelin receptor A

Direct thrombin inhibition reduces lung collagen, accumulation, and connective tissue growth factor mRNA levels in bleomycin-induced pulmonary fibrosis

The multivalent activity of the tissue factor-thrombin pathway in thrombotic and non-thrombotic disorders as a target for therapeutic intervention

NF-κB Pathway as a Potential Target for Treatment of Critical Stage COVID-19 Patients

Interleukin-6 Is a Biomarker for the Development of Fatal Severe Acute Respiratory Syndrome Coronavirus 2 Pneumonia

A molecular single-cell lung atlas of lethal COVID-19

Activation of NF-κB and induction of proinflammatory cytokine expressions mediated by ORF7a protein of SARS-CoV-2

SARS-CoV-2 drives JAK1/2-dependent local complement hyperactivation

Granulocyte colony-stimulating factor (G-CSF) increases histone-complexed DNA plasma levels in healthy volunteers

Self-sustaining IL-8 loops drive a prothrombotic neutrophil phenotype in severe COVID-19

Patients with COVID-19: in the dark-NETs of neutrophils

NETopathies? Unraveling the Dark Side of Old Diseases through Neutrophils, Front Immunol

IL-6 in inflammation, immunity, and disease

Role of JAK/STAT in Interstitial Lung Diseases; Molecular and Cellular Mechanisms

COVID-19 pulmonary pathology: a multi-institutional autopsy cohort from Italy and

The Emerging Threat of (Micro)Thrombosis in COVID-19 and Its Therapeutic Implications

Antithrombin III deficiency-induced coagulopathy in the context of COVID-19: a case series

Association between antithrombin and mortality in patients with COVID-19. A possible link with obesity

Anticoagulation with argatroban in patients with acute antithrombin deficiency in severe COVID-19

Plasma Antithrombin Values Are Significantly Decreased in Coronavirus Disease 2019 (COVID-19) Patients with Severe Illness

Extracellular DNA traps promote thrombosis

Inferring molecular mechanisms of dexamethasone therapy in severe COVID-19 from existing transcriptomic data

Dexamethasone modulates immature neutrophils and interferon programming in severe COVID-19

Effect of anakinra versus usual care in adults in hospital with COVID-19 and mild-to-moderate pneumonia (CORIMUNO-ANA-1): a randomised controlled trial

Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial

Targeting JAK-STAT Signaling to Control Cytokine Release Syndrome in COVID-19

JAK-STAT Pathway Inhibition and their Implications in COVID-19 Therapy

Efficacy and safety of baricitinib for the treatment of hospitalised adults with COVID-19 (COV-BARRIER): a randomised, double-blind, parallel-group

COVID-19: combining antiviral and anti-inflammatory treatments

Dysregulated adaptive immune response contributes to severe COVID-19

COVID-19 Lung Pathogenesis in SARS-CoV-2 Autopsy Cases

SARS-CoV-2 proteins regulate inflammatory, thrombotic and diabetic responses in human arterial fibroblasts

Angiotensin II triggers release of neutrophil extracellular traps, linking thromboinflammation with essential hypertension

Cancer-associated fibroblastsderived exosomes upregulate microRNA-135b-5p to promote colorectal cancer cell growth and angiogenesis by inhibiting thioredoxin-interacting protein

MiR155-5p in adventitial fibroblasts-derived extracellular vesicles inhibits vascular smooth muscle cell proliferation via suppressing angiotensin-converting enzyme expression

Tissue factor expression in neutrophil extracellular traps and neutrophil derived microparticles in antineutrophil cytoplasmic antibody associated vasculitis may promote thromboinflammation and the thrombophilic state associated with the disease