Therapeutic agents for ARDS

2.1 ACE2

The renin-angiotensin system (RAS) is involved in ARDS pathophysiology. Patients with ARDS present increased levels of Angiotensin II, a vasoconstrictor involved in inflammation and pulmonary edema that exerts its activities through angiotensin type I receptor [9]. Angiotensin Converting enzyme II (ACE2) hydrolyses Angiotensin II producing Angiotensin 1–7, which has been found to be protective in experimental models. In a randomized phase 2a clinical study, GSK2586881 (recombinant human ACE2) was administered as an exogenous ACE2, in order to hydrolize Angiotensin II, and proved safety but did not improve clinical outcomes in ARDS patients requiring mechanical ventilation [10].

In patients with COVID-19, SARS-CoV-2 is known to bind ACE2; both membrane-bound (mACE2) and soluble (sACE2) forms. However, only mACE2 mediates the virus entrance into the cell, but not sACE2. Angiotensin type I receptor blockers increase the levels of Angiotensin II, which stimulates ACE2 shedding; sACE2 catalyzes the conversion of Angiotensin II to Antiotensin 1–7 while also binds SARS-CoV-2 blocking its entrance to the host cells. Presently, there is an ongoing randomized phase 2 trial with oral 50 mg Losartan (an angiotensin type I receptor blocker) and 25 mg Spironolactone (a blocker of aldosterone secretion) in patients with COVID-19-ARDS (NCT04643619).

2.2 Alveolar epithelium

The alveolar epithelium has a key role in ARDS severity [11]. It is composed of alveolar type I cells (ATI cells), which cover the 95% of alveolar surface and are the major responsible of gas exchange, and alveolar type II cells (ATII cells), which are the progenitor cells of the alveolar epithelium and can proliferate and differentiate into ATII cells. ATII cells also produce surfactant. Both cell types are critical in ion transport and present immunologic functions [12, 13].

2.3 Alveolar endothelium

The alveolar endothelium is exposed to higher oxygen tensions while maintaining low-pressure blood flow compared to the systemic vascular endothelium. When there is a damage, injured alveolar endothelium promotes the destruction of the vascular bed and the expression of proinflammatory, reactive oxygen species and recruitment molecules, together with enhanced procoagulant activity and clot formation [[14].

5.1 Neuromuscular blockers

The neuromuscular transmission is blocked by neuromuscular blocking agents at the neuromuscular junction, in order to minimize volutrauma, ventilator-induced lung injury, and biotrauma [49].

In a multicenter randomized trial, the early administration of neuromuscular-blocking agent cisatracurium in patients with moderate to severe ARDS improved 90-day survival and the time off the ventilator [50]. However, in another clinical trial, and early and continuous infusion of cisatracurim did not decrease 90-day mortality in patients with moderate-to-severe ARDS [51]. Current evidence favors avoiding a continuous infusion of neuromuscular blockers in patients with mechanical ventilation but use a lighter sedation strategy, and for patients who need a deep sedation to facilitate lung protective ventilation or prone positioning, to infuse neuromuscular blockers for 48 h is a reasonable option [52].

5.2 Steroids

Steroids are powerful anti-inflammatory and anti-fibrotic drugs that may lead to high-risk infections due to the suppression they exert on the immune system.

Clinical trials suggest that steroid treatment in ARDS patients would be indicated at the onset of the pathology. Administered corticosteroids 72 h after ARDS diagnosis decreased lung damage and increased ventilator weaning [53]. A meta-analysis in ARDS patients concluded that low-dose corticosteroids in early ARDS significantly reduced mortality and the duration of mechanical ventilation, whereas high doses did not [54]. A different meta-analysis shows that steroid treatment improves mortality and promotes shorter ventilation periods [55]. In contrast, in patients with influenza pneumonia, the early use of steroid therapy is associated with increased mortality [56, 57]. Nonetheless, studies in patients with community-acquired pneumonia treated with corticosteroids showed a reduced risk of treatment failure [58], reduced mortality, hospital stay and need for mechanical ventilation [59].

Dexamethasone is one of the most clinically used steroids for treatment. In a phase 2/3 trial patients with moderate-to-severe ARDS requiring mechanical ventilation were intravenously administered with dexamethasone (20 mg for 5 days, then 10 mg for the next 5 days) and presented an increase in the number of ventilator-free days and reduced mortality [60].

In COVID-19 patients, treatment with dexamethasone (intravenous or oral, 6 mg/day for 10 days) resulted in a lower incidence of death in those patients requiring invasive mechanical ventilation compared to those not requiring ventilator support [61]. Some of the clinical trials now recruiting are the phase 4 REMED study (NCT04663555), which aims to test two different doses (6 mg vs. 20 mg) of intravenous dexamethasone in SARS-CoV-2-induced ARDS patients. Or a phase 3 study that aims to compare intravenous treatment with dexamethasone or methylprednisolone in COVID-19 patients with ARDS (NCT04499313).

Regarding hydrocortisone, in a trial, patients with ARDS-associated sepsis were treated with a dose of 50 mg every 6 h within 12 h of their ARDS diagnosis. The treated group showed improvements in pulmonary physiology, but not a decrease in mortality compared to the placebo group [62].

Another of the most investigated corticosteroids for future therapies is methylprednisolone. In 24 patients with severe ARDS methylprednisolone (2 mg/kg/day for 32 days) decreased in-hospital and ICU mortality [63]. In the first 72 h, patients with ARDS were treated with an infusion of methylprednisolone (1 mg/kg/day) for 28 days, and had decreased C-reactive protein, mechanical ventilation and mortality [64]. A phase 2 study proposed intrapleural administration of the steroid Solumedrol (methylprednisolone) versus conventional treatment with extracorporeal membrane oxygenation and intravenous steroid administration. Results are not yet available (NCT01423864).

The MINECRAFT phase 2 study, aims to study the efficacy of administering canrenone, a steroidal antimineralocorticoid, intravenously in moderate-to-severe ARDS patients due to SARS-CoV-2 infection (NCT04977960).

A recently explored field is the administration of inhaled steroids. Early treatment consisting of inhaled budesonide together with a beta-agonist in patients at risk of developing ARDS improved oxygenation [65]. Another study where nebulised budesonide was administered also improved oxygenation and reduced proinflammatory cytokines (Tumor necrosis factor-$\alpha$ (TNF-$\alpha$), Interleukin (IL)-1$\beta$ and IL-6) [66]. There is a phase 2 study in paediatric ARDS patients with inhaled budesonide (NCT04064684). In neonatal patients with severe ARDS requiring mechanical ventilation, intratracheal treatment with budesonide and surfactant resulted in a decreased incidence of bronchial dysplasia or death and decreased inflammation [67]. In children on mechanical respiratory support, treatment with budesonide and surfactant did not improve survival or the development of bronchial dysplasia over the surfactant-treated group but decreased the need for mechanical ventilation [68].

5.3 Statins

Statins are $\beta$-Hydroxy $\beta$-methylglutaryl-CoA (HMG-CoA) reductase inhibitors with immunomodulatory properties. A meta-analysis showed that treatment with statins prior to intensive care unit (ICU) admission or before a diagnosis of a specific pathology showed a decrease in 30-day mortality, but no association with in-hospital mortality [69].

The Hydroxymethylglutaryl-CoA reductase inhibition with simvastatin in Acute lung injury to Reduce Pulmonary dysfunction (HARP-2) trial was a multicentre trial that sought to test simvastatin (80 mg/day) in ICU patients, 48 h after the onset of ARDS. Patients involved in the study could be divided into two different sub-phenotypes: hypo-inflammatory (65%) and hyper-inflammatory (35%), and only increased survival was found in patients who had a hyper-inflammatory sub-phenotype treated with simvastatin. This study highlighted the need to phenotype different types of ARDS patients [70]. In the Statins for Acutely Injured Lungs from Sepsis (SAILS) trial they were also able to identify different biological phenotypes but did not see phenotype-specific benefit from rosuvastatin treatment [71]. In recent years, clinical studies propose to investigate the role of statins in ARDS patients of different aetiologies, although those have been cancelled due to lack of enrolment or other causes.

5.4 Carbon monoxide

Carbon monoxide (CO) results from the catabolism of heme oxygenase within the body. Its anti-inflammatory and anti-apoptotic role has been described. CO down-regulates the NOD-like receptor family pyrin domain containing 3 (NLRP3) inflammasome, thus preventing mitochondrial dysfunction, and protects against cellular oxidative stress in models of lung injury [72, 73]. In in vivo models that received lypopolysaccharide (LPS), 50 parts per million (ppm) inhaled CO restored arterial resistance and decreased NOS-2 expression, although no changes were seen in plasma levels of inflammatory cytokines [74]. In a nonhuman primate pneumonia model CO treatment (200 ppm of concentration for 60 minutes) reduced extravascular alveolar fluid [75].

In a phase 1 trial in patients with ARDS-induced sepsis, a low dose (100–200 ppm) of inhaled CO was found to be a well-tolerated and safe treatment during mechanical ventilation. A phase 2 trial is currently recruiting ARDS patients to be treated with inhaled carbon monoxide at 200 ppm (NCT03799874).

5.5 Mesenchymal Stromal Cells

Mesenchymal Stromal Cells (MSCs) have immunomodulatory properties and reparative effects on damaged tissue, presenting paracrine and cell-cell communication (see chapter Cell Therapies in ARDS). Their role as a treatment in ARDS depends on the microenvironment to which the cell therapy is exposed and what has caused the lung injury [76].

The phase 1/2 clinical trial MultiStem Therapy in ARDS (MUST-ARDS) evaluated the safety of intravenous 900 million bone marrow-derived multipotent adult progenitor cells administered within 96 h of the onset of moderate-to-severe ARDS patients requiring mechanical ventilation. Administration of the cells was well tolerated and tended to decrease the need for mechanical ventilation [77].

In the Human Mesenchymal Stromal Cells for ARDS (START) phase 2a trial, patients with moderate-to-severe ARDS requiring mechanical ventilation were given an intravenous dose of MSCs, which was safe but showed no improvement over the placebo-treated group. These findings have been attributed to the low viability of the administered cells [78]. The Mesenchymal Stromal Cells for ARDS (STAT) phase 2b trial, an extension currently recruiting, aims to test the safety and efficacy of 10 million MSCs/kg (NCT03818854).

The REALIST trial proposes to investigate whether a single infusion of MSCs (human umbilical cord-derived CD362 enriched MSCs) could help in the treatment of ARDS, a phase 1/2 study (NCT03042143).

Regarding COVID-19-induced ARDS, it has been shown that in 7 patients who received a transfusion of ACE2${}^{-}$ MSCs, lung function and symptomatology improved two days after treatment, and inflammation was reduced by decreasing C reactive protein (CRP) and TNF-$\alpha$[79].

5.6 Regulatory T-cells

Regulatory T-cells (Treg cells) act on the immune system by decreasing its activation and promoting homeostasis. Overexpression of Transforming Growth Factor (TGF) $\beta$1, the most secreted cytokine by Treg cells, in a murine model of acute lung injury (ALI) induces more Treg cells and decreases T helper 17 cells (Th17) cells, improving lung inflammation [80].

Several clinical trials are currently ongoing in COVID-19-ARDS patients proposing intravenous administration of Treg cells (NCT05027815 and NCT04468971), and a study in COVID-19-ARDS patients receiving intravenous Treg/Th2 hybrid cells has just been terminated, although results are not posted yet (NCT04482699).

5.7 Vitamin C

Vitamin C is an antioxidant molecule with protective effects. In one study, vitamin C levels were found to be undetectable in more than 90% of patients with SARS-CoV-2-associated ARDS [81]. In the Vitamin C in patients with Sepsis and Severe Acute Respiratory Failure (CITRIS-ALI) phase 2 trial, patients with sepsis and consequent ARDS were treated with a 96h-infusion of vitamin C. There was no difference between the vitamin C-treated group and the control group in terms of decreased inflammation, but secondary outcomes showed a decrease in 28-day mortality in the treated group [82].

Completed but unpublished clinical studies include COVID-19 patients with ARDS treated with vitamin C and other antioxidants (NCT04570254), and ascorbic acid (NCT04710329). Also, there is a phase 3 study in septic patients with ARDS that proposes to compare the effect of high-dose intravenous vitamin C, but is not yet enrolling patients (NCT04404387).

5.8 Ulinastatin

Ulinastatin, a glycoprotein known as urinary trypsin inhibitor, is an experimental drug with anti-inflammatory properties. A clinical study of 14 consecutive days of treatment with ulinastatin in ARDS patients requiring mechanical ventilation resulted in decreased TNF-$\alpha$, IL-6 and CRP levels, increased antioxidant capacity, decreased ventilatory need and hospital-stay days [83].

5.9 Inhibitors

6.1 Tissue Factor

Tissue Factor (TF) is a transmembrane protein that is the major initiator of the extrinsic coagulation pathway when activated by the binding of factor VIIa. ALT-836 is an anti-TF antibody that blocks the binding of factor VIIa, and a single intravenous dose of ALT-836 (0.06, 0.08, 0.1 mg/kg) has proved to be safe in a randomized controlled phase 1 trial in patients with ARDS requiring mechanical ventilation [123]. A randomized phase 2 clinical trial in patients with sepsis and ARDS receiving a single (0.06 mg/kg) intravenous dose up to four doses has already been performed, although results have not been published yet (NCT00879606).

Tissue Factor Pathway Inhibitor (TFPI) modulates the initiation of the extrinsic coagulation pathway. A randomized controlled phase 3 clinical trial of intravenous tifacogin (recombinant TFPI) administration during 96 h in patients with severe sepsis did not reduce mortality and was associated with increased bleeding [124]. However, in a randomized controlled phase 3 clinical trial intravenous tifacogin administration during 96 h did not decrease mortality but reduced prothrombin fragment and thrombin antithrombin complexes levels in patients with severe community-acquired pneumonia [125].

6.2 Antithrombin

Antithrombin is a serine protease inhibitor synthesized in the liver [126, 127], and is known to inhibit procoagulant enzymes including thrombin, factor Xa, IXa, XIa and XIIa. When heparin binds to antithrombin, its inhibitory activity is 1000-fold increased [127].

A randomized phase 3 clinical trial with intravenous 30,000 IU of antithrombin within 4 days had no effect on mortality in patients with severe sepsis (the KyberSept Trial), although an increased risk of hemorrage was detected when administering antithrombin and heparin together [128]. A post hoc analysis in patients with severe sepsis treated in a single center early after onset revealed increased bleeding due to antithrombin [129].

Nebulized antithrombin alone or combined with heparin attenuated lung injury in HCl/LPS-induced ALI in rats, reducing pulmonary coagulation and inflammation without altering systemic coagulation nor bleeding [130].

6.3 Heparin

Heparin is a natural anticoagulant produced by mast cells in the intestine or lungs, basophils in the blood and endothelial cells [131]. Heparin presents anticoagulant actions potentiating antithrombin inhibitory activity and enhancing TFPI, and anti-inflammatory actions both related or not to thrombin inhibition [132].

Controversial results have been determined in patients with ARDS while administering local heparin. In a phase 1 trial nebulized heparin (50,000 IU/day, 100,000 IU/day, 200,000 IU/day, 400,000 IU/day) did not produce adverse effects and attenuated pulmonary coagulopathy in patients with ARDS requiring mechanical ventilation [133, 134], and in a randomized phase 2 study nebulized heparin (25,000 IU) reduced days of mechanical ventilation in patients with ARDS [135]. Also, in a randomized phase 3 clinical trial (CHARLI) nebulized heparin (250,000 IU) every 6 h to day 10 was well tolerated with decreased lung injury progression and earlier return at home in patients with invasive ventilation [136]. In contrast, in a randomized controlled trial with nebulized heparin focused on the safety and efficacy of burn patients with inhalation trauma (HEPBURN), the trial was stopped because of increased systemic clotting times and adverse events [137].

Nebulized heparin has also been proposed for COVID-19 patients. A randomized phase 2/3 clinical trial with nebulized 25,000 IU of heparin every 6 h for up to 10 days in patients with COVID-19 requiring mechanical ventilation is being performed (NCT04545541).

6.4 Thrombomodulin

The Protein C Pathway also has a major role in coagulation and fibrinolysis regulation. Thrombomodulin is a thrombin receptor, and, when the complex is formed, protein C is cleaved and activated protein C is produced.

ART-123 is a recombinant human soluble thrombomodulin. In a randomized controlled phase 2b study intravenously administered ART-123 for 6 days proved to be safe and effective reducing prothrombin fragment and thrombin-antithrombin complex concentrations in patients with sepsis-associated disseminated intravascular coagulation [138]. In addition, in a retrospective study intravenously combined sivelestat and recombinant human soluble thrombomodulin improved 60-day survival and ventilator-free days in patients with ARDS and disseminated intravascular coagulation [139]. In the full analysis of the phase 3 multinational Scarlet study that evaluate the efficacy and safety of intravenous ART-123 during 6 days to treat sepsis-associated coagulopathy, no statistically differences were determined [140]. However, in post hoc analysis in patients with sepsis-associated coagulopathy that did not receive concomitant heparin, ART-123 proved more benefit, indicating that heparin administration could impact ART-123 efficacy, a fact that should be confirmed in further studies [141].

6.5 Activated Protein C

Alveolar epithelial cells release thrombomodulin from the cell surface, due to a metalloproteolytic process, and this decreases the ability of these cells to activate Protein C [120, 121].

Inhaled drotrecogin alfa (recombinant human activated Protein C) in patients with ARDS decreased coagulation, neutrophils recruitment and inflammation in the alveolar compartment and increased fibrinolysis without producing systemic effects [142]. However, in a randomized multicentre phase 3 Prowess-Shock trial intravenous drotrecogin alfa (recombinant human activated Protein C) in 1967 patients with septic shock did not reduce mortality [143], and no further studies have been performed becaause Activated Protein C was removed from the market.

6.6 Streptokinase

Plasminogen activator and inhibitor pathway regulate fibrin deposition. Streptokinase binds plasminogen and drives the conversion of plasminogen to plasmin, a fibrinolytic enzyme. In a randomized phase 3 trial nebulized streptokinase in patients with severe ARDS improved oxygenation and lung mechanics [144].

8.1 Alpha-1 antitrypsin

Alpha-1 antitrypsin is a serine protease inhibitor that has been found to ameliorate oxygenation [146]. A randomized phase 2 clinical trial with intravenous prolastin (plasma-purified alpha-1 antitrypsin) in patients with COVID-19 ARDS is presently being conducted [147].

8.2 Transient Receptor Potential Vanilloid 4 inhibitor

The mechanosensitive cation calcium channel Transient Receptor Potential Vanilloid 4 inhibitor (TRPV4) is an essential homeostasis regulator that is implicated in ARDS inflammation [148]. TRPV4 can induce alveolar endothelial and epithelial dysfunction, which results in increased permeability and edema [149]. Nevertheless, in a preclinical model with intratracheal Pseudomonas aeruginosa in mice, TRPV4 activity has demonstrated to enhance macrophages phagocytosis and decrease inflammation [150]. Various TRPV4 inhibitors have proved to decrease acute lung injury in preclinical models. In mice exposed to hydrochloric acid or chlorine gas, TRPV4 inhibitor reduced inflammation and vascular leakage [151]. In a first clinical study, TRPV4 inhibitor did not produce ameliorations in healthy patients receiving inhaled LPS (NCT03511105).

8.3 Matrix metalloproteinase 12 inhibitor

FP-025 inhibits matrix metalloproteinase-12 (MMP12), an enzyme that degrades and remodels the extracellular matrix but is also known to modulate the influx of monocytes and macrophages in the alveolar compartment [152]. There is an ongoing randomized phase 2/3 clinical study with FP-025 (100 or 300 mg) in patients with severe and critical COVID-19-ARDS (NCT04750278).

8.4 Sevoflurane

Sedation with the volatile anesthetic sevoflurane-induced anti-inflammatory processes in ventilated patients [153]. In a randomized controlled phase 3 clinical trial volatile or intravenous sedation with sevoflurane for 48 h has been administered in patients with COVID-19-ARDS, although results have not been published yet (NCT04355962). Another randomized phase 3 trial to determine the effects of inhaled sevoflurane sedation on extravascular lung water and pulmonary vascular permeability in ARDS patients is proposed, although the recruitment has not started (NCT04530188).

9.1 Adenosine A2A receptor agonists

The nucleoside adenosine has anti-inflammatory properties, and its deficiency has been shown to increase pulmonary oedema and inflammation in a murine model of VILI [154].

Pharmacological intervention with the adenosine A2A receptor agonist CGS-21680 in rat VILI models reduced pulmonary edema, respiratory elastance and neutrophil recruitment into the lung compared to vehicle-treated animals [155]. In different models of ALI induced by HCl, LPS or Escherichia coli, instillation of the agonist GW328267C led to alveolar fluid clearance [156].

Blockade of equilibrative nucleoside transporters (ENTs) with dipyridamole increases adenosine in the alveolus and decreases pulmonary edema and improves gas exchange during ALI [157, 158].

9.2 Protease-activated receptor 1

Coagulation activates inflammation through protease-activated receptors (PARs) [159]. PAR1 is expressed in epithelial lung cells and the endothelium and is associated with a prothrombotic state [160, 161]. Thrombin binds to PAR1 and stimulates neutrophil recruitment and the release of proinflammatory cytokines [162]. Nevertheless, in antigen-presenting cells, PAR1 activation decreases the production of proinflammatory cytokines [163].

In influenza virus infection in mice, after activating PAR1 receptors with an agonist, they found increased lung inflammation but did not affect survival. They also observed that activated PAR1 increased the conversion of plasminogen to plasmin [164]. PAR1 antagonists are only in clinical trials for other pathologies.

9.3 Receptor for advanced glycation end-products inhibitors

The soluble receptor for advanced glycation end-products (sRAGE) is a marker of epithelial damage, especially in ATI, and is a prognostic marker for ARDS [165, 166].

Blockade of RAGE, using anti-RAGE antibody or sRAGE decoy receptor in acid-induced mice model of ALI reduced RAGE mRNA levels in the lung, restored alveolar-capillary barrier permeability after injury, decreased the total number of leukocytes in bronchoalveolar lavage (BAL) and restored membrane aquaporin-5 expression [167]. In piglets, blockade of RAGE decreased alveolar inflammation and induced alveolar fluid clearance [168]. A recent study in LPS-induced ALI murine model has shown that RAGE signalling mediates epithelial barrier dysfunction, enhancing lung inflammation and causing loss of adherent junctions [169].

9.4 Haptoglobin

Haptoglobin acts as a scavenger receptor for cell-free haemoglobin (CFH), which is elevated during sepsis and correlated with increased mortality. Haptoglobin decreases CFH levels and iron levels, leading to less oxidative damage to the lung in sepsis, but it has not been shown to reduce inflammation [170]. In a transgenic mouse overexpressing haptoglobin in alveolar macrophages, CFH clearance and decreased lung injury were observed, suggesting that haemoglobin catabolism is linked to iron mobilisation in macrophages [171].

9.5 Lipoxin A4

Lipoxin A4 (LXA4) is derived from arachidonic acid that promotes alveolar epithelial wound repair and the proliferation and differentiation of ATII cells into ATI cells [172]. In vivo, LPS-induced ALI model animals resulted in decreased levels of TNF-$\alpha$ and IL-1$\beta$, inhibition of neutrophil recruitment to the lung, inhibition of ATII cell apoptosis and epithelial-mesenchymal transition [173, 174].

Resolvin D1 is a specialised pro-resolving mediator that acts by stimulating the lipoxin A4 receptor on immune cells, reducing ROS generation, blocking nuclear factor kappa-light-chain-enhancer of activated B cells (NF-$\alpha$$\beta$) activation and accelerating the production of antioxidant proteins. In a mouse model, resolvin D1 treatment reduces leukocyte infiltration and inflammatory cytokine release [175, 176]. In a rat ALI model, resolvin administration attenuated LPS-induced ALI and promoted alveolar fluid clearance by increasing the expression of sodium and Na, K-ATPase channels [177].