Host-Based Treatments for Severe COVID-19

COVID-19 has been a global health problem since 2020. There are different spectrums of manifestation of this disease, ranging from asymptomatic to extremely severe forms requiring admission to intensive care units and life-support therapies, mainly due to severe pneumonia. The progressive understanding of this disease has allowed researchers and clinicians to implement different therapeutic alternatives, depending on both the severity of clinical involvement and the causative molecular mechanism that has been progressively explored. In this review, we analysed the main therapeutic options available to date based on modulating the host inflammatory response to SARS-CoV-2 infection in patients with severe and critical illness. Although current guidelines are moving toward a personalised treatment approach titrated on the timing of presentation, disease severity, and laboratory parameters, future research is needed to identify additional biomarkers that can anticipate the disease course and guide targeted interventions on an individual basis.


Introduction
The SARS-CoV-2 coronavirus was first identified as a cause of severe pneumonia in Wuhan, but it quickly spread worldwide, and on 11 March 2020, the World Health Organization (WHO) declared COVID-19 a global pandemic [1]. Currently, more than 750 million confirmed cases of COVID-19 have been reported globally. Currently, COVID-19 infection still has a significant burden on the global health care system, and more than 6 million deaths have been reported [2].
The manifestations of SARS-CoV-2 infection range from asymptomatic or mild forms of flu-like illness to severe forms of bilateral interstitial pneumonia resulting in acute respiratory failure requiring maintenance therapies and admission to an intensive care unit.
The WHO has classified COVID-19 as mild, moderate, severe, and critical illness. Mild illness consists of symptoms of SARS-CoV-2 infection (such as fever, cough, fatigue, wheezing), but without manifestations of viral pneumonia or hypoxia. Moderate, severe, and critical illness share common evidence of pneumonia, with an increasing severity of lung involvement and then progressively worsening clinical manifestations. Patients with moderate COVID-19 are defined by clinical signs of pneumonia with SpO2 ≥ 90% with

Corticosteroids
Corticosteroids are recommended by the current WHO guidelines as first-line therapy in hospitalised patients with severe and critical management of COVID-19 [4,15].
Glucocorticoids (GCs) are endogenous modulators of systemic inflammation in addition to exerting a number of other genomic and nongenomic effects. Indeed, cytokines secreted during the innate immune response (e.g., IL1-b and IL-6) activate the hypothalamicpituitary-adrenal axis to secrete GCs, which exert their immunomodulatory function by binding the ubiquitously expressed glucocorticoid receptor-α (GR-α). When there is a massive inflammatory response, such as in severe COVID-19, glucocorticoid-mediated anti-inflammatory activity may be inadequate for the severity of the patient's critical illness (corticosteroid-related critical illness failure, CIRCI) [16,17].
Glucocorticoids had already been shown to be effective in reducing mortality in sepsis, severe pneumonia, or ARDS from any cause.
Several studies have examined the use of GCs in community-acquired pneumonia (CAP) on the assumption that the immunomodulating effect of these drugs was able to control inflammation and thus promote the healing process and avoid complications, particularly those related to sepsis, by reducing lung and systemic inflammation, especially in severe CAP [18,19]. The extensive literature related to this therapeutic approach has been analysed over time in several reviews and meta-analyses that have confirmed a more rapid improvement in the clinical parameters of severe forms of CAP treated with GCs, but without always reflecting a decrease in mortality or ICU stays [20,21]; in fact, the American Thoracic Society (ATS) guidelines emphasise that the use of GCs in severe forms of CAP is still partially controversial and requires further research in this area [22]. In order to elucidate a reasoned use of steroid therapy in patients with ARDS, in 2020, Meduri et al. reviewed the pharmacological principles underlying the use of GCs, which should guide both the timing of the use of these drugs and their dosing and tapering; they also suggested the clinical and laboratory parameters that should be monitored during therapy, highlighting how steroid therapy should be individualised for each patient in order to have a successful outcome [23].
Based on these assumptions and a multicentre, retrospective cohort study by the Saudi Critical Care Trial Group that advised against the use of GCs in critically ill patients with Middle East Respiratory Syndrome (MERS) [24], the WHO initially did not recommend the use of steroid therapy for the treatment of SARS-CoV-2 infections [25].
Despite initial scepticism about the use of GCs in COVID-19 disease, the extensive literature reviewed in this critical review subsequently validated the immunomodulatory effect of these drugs and confirmed the rationale for their use in routine clinical practice.
In June 2020, the RECOVERY clinical trial showed for the first time, in a prospective randomised trial compared with placebo, that treatment with 6 mg dexamethasone for up to 10 days reduced the risk of death at 28 days in patients hospitalised with COVID-19 undergoing mechanical ventilation or oxygen therapy, with a greater impact among patients hospitalised with COVID-19 receiving respiratory support with mechanical ventilation (36% reduction) compared with oxygen therapy (18% reduction). In contrast, GCs showed an increased risk of death among patients who did not require respiratory support at the time of randomisation [26].
Since the RECOVERY trial, subsequent clinical trials have focused on patients with severe COVID-19.
The multicentre, randomised, open-label, CODEX clinical trial, conducted in 41 intensive care units (ICUs), tested the efficacy of intravenous dexamethasone plus standard of care versus placebo on ventilator-free and ventilator-free days in patients with moderate and severe ARDS due to SARS-CoV-2 infection. CODEX confirmed a statistically significant increase in the number of ventilator-free days over 28 days in patients treated with dexamethasone (6.6 ventilator-free days versus 4.0 in the standard care group) [27].
The COVID-STEROID 2 study group proposed treatment with higher doses of dexamethasone. This multicentre clinical trial randomised 1,000 patients with SARS-CoV-2 pneumonia and severe hypoxemia in a 1:1 ratio to receive either 12 or 6 mg of intravenous dexamethasone for up to 10 days and found no statistically significant differences by treating patients with a higher dose of steroids. In fact, the two groups showed a similar number of days of life without life support at 28 days [28].
The REMAP-CAP study proposed the use of an alternative steroid molecule to dexamethasone: 384 patients with confirmed SARS-CoV-2 infection admitted to the ICU were randomised to receive a fixed dose of hydrocortisone (50 or 100 mg intravenously) or a shock-dependent dose of hydrocortisone compared with standard of care alone.
This study confirmed the superiority of steroid therapy with hydrocortisone over the 2020 standard of care, showing an improvement in organ support-free days within 21 days [29].
Dequin et al. also analysed the effect of intravenous hydrocortisone for 7 days at higher doses (200 mg) in a multicentre randomised double-blind sequential trial (CAPE COVID) involving 149 critically ill patients with SARS-CoV-2 pneumonia. This study was stopped early but showed no improvement in organ support-free days within 21 days and showed no statistically significant differences between the placebo and steroid groups in treatment failure (defined as death or persistent respiratory support) at day 21 [30].
Finally, several clinical trials have proposed a treatment protocol with methylprednisolone.
Edalatifard et al. proposed three-day treatment with pulsed intravenous methylprednisolone (250 mg daily) versus placebo in a randomised controlled, single-blind, two-arm, parallel clinical trial. This study included 68 patients randomised in a 1:1 ratio with COVID-19 infection, abnormal computed tomography (CT) findings, and oxygen saturation (SpO2) < 90% at rest. Patients treated with methylprednisolone showed a lower mortality rate than the standard of care (5.9% vs. 42.9%). In addition, patients in the methyl-prednisolone group who required mechanical ventilation or high-flow oxygen (HFO) or low-dose oxygen via nasal canula were characterised by a lower incidence of death (7.7%, 8.3%, and 0%, respectively, compared with 60%, 57.1%, and 22% in the standard group) [31].
Metcovid was a parallel, double-blind, placebo-controlled, randomised, phase IIb clinical trial that proposed a lower dose of intravenous methylprednisolone (0.5 mg/kg) for a longer period (5 days) and included a larger cohort of patients. A total of 416 patients were randomised to receive methylprednisolone versus standard of care in a 1:1 ratio. The inclusion criteria included clinical and radiological signs of COVID-19, SpO2 ≤ 94% in room air, or the need for oxygen support (either low-dose or high-dose oxygen, invasive or non-invasive mechanical ventilation). Metcovid showed no statistically significant differences in 28-day mortality between the two groups, suggesting that 5 days of low-dose methylprednisolone did not change the patients' prognosis [32].
In a multicentre, observational, longitudinal study, Salton et al. evaluated the efficacy of a protocol of infusion of 80 mg methylprednisolone daily for at least 14 days and until clinical improvement [33] in 83 hospitalised patients with confirmed SARS-CoV-2-related ARDS, PaO2/FiO2 < 250 mmHg, and CRP > 100 mg/L compared with 90 unexposed controls who met the same inclusion criteria. In this study, 28-day mortality was significantly reduced in the methylprednisolone-treated group (adjusted HR, 0.29) along with increased days free of mechanical ventilation and shorter median duration of invasive mechanical ventilation among patients admitted to the ICU [34]. Despite the methodological limitations arising from the observational design, this study was the first to investigate a titration of GC treatment duration on clinical status, assessed by a reduction in CRP and an increase in PaO2/FiO2. This is in agreement with the previous 2017 task force proposal of the Society of Critical Care Medicine and the European Society of Intensive Care Medicine for the management of critically ill patients with sepsis and septic shock, acute respiratory distress syndrome, and major trauma [35].
Overall, the REACT working group published a meta-analysis of seven clinical trials that confirmed the reduction in 28-day mortality with steroid use in patients with COVID-19 severe pneumonia. The clinical trials analysed by the REACT study examined different molecules: low-dose and high-dose dexamethasone, low-dose hydrocortisone, and highdose methylprednisolone [36].
Thus, once corticosteroid therapy for severe COVID infections was validated, the clinicians focused on exploring the most appropriate molecules, dosage, and timing of administration.
Ranjbar et al. compared dexamethasone and methylprednisolone in a prospective randomised, triple-blind controlled trial, assigning 86 hospitalised COVID-19s into two groups treated with higher doses of methylprednisolone (2 mg/kg/day) or dexamethasone (6 mg/day) for 10 days. Overall, the group treated with methylprednisolone showed better results than the group treated with dexamethasone. In particular, patients treated with methylprednisolone had significantly better clinical status at days 5 and 10. In addition, treatment with methylprednisolone resulted in a shorter length of hospital stay and less need for ventilatory support (18.2% compared with 38.1% in the dexamethasone group) [37].
The same drug protocols were analysed by Saeed et al. in a prospective cohort study of 414 COVID-19 patients with ARDS, according to Berlin definition [3], and who were mechanically ventilated. Patients treated with methylprednisolone showed a significant reduction in ICU length of stay (7.33 in the methylprednisolone group vs. 19.43 + 4.42 in the dexamethasone group). This study also analysed the effect of the two different pharmacological protocols on biological markers: the two groups of patients showed similar blood levels of C-reactive protein (CRP), D-dimer, cytokines, and LDH at the time of ICU admission, while the same inflammatory markers were significantly lower in the methylprednisolone group after 10 days [38].
These studies seem to suggest the use of methylprednisolone in subgroups of patients with severe SARS-CoV-2 pneumonia; however, further analysis of larger samples is needed to generalise these findings.
The Italian randomised controlled open-label clinical trial (MEDEAS) examined a protocol of infusional methylprednisolone at an initial dose of 80 mg/day for 7 days, and then gradually reduced every 3 days only if the patient had a PaO2:FiO2 > 200 compared with dexamethasone 6 mg/day for 10 days or until hospital discharge (whichever was earlier) in 677 participants, randomised into two parallel arms with a 1:1 allocation ratio, showing no statistically significant differences in 28-day mortality in the two arms. Despite the non-superiority of one drug over the other, MEDEAS also showed that the arm of patients treated with methylprednisolone was characterised by longer mean hospitalisation times, partly because of the inherent duration of the treatment itself, but by a lower incidence of intensive care unit admissions. This is, to date, the only multicentre, openlabel, randomised controlled clinical trial comparing the two treatment protocols and which considered the largest number of patients, involving 26 centres throughout Italy [39]. In conclusion, the current WHO guidelines suggest the use of any GC molecule, at a dose equivalent to 6 mg dexamethasone for 7-10 days, in moderate and severe COVID-19 [4]. This treatment regimen has been shown to be safe and effective in several randomised clinical trials, reducing all-cause mortality without being associated with a higher incidence of complications or adverse events than usual care. Despite its unquestionable efficacy, it is likely that this protocol could be further improved. Indeed, it is still unclear which GCs molecule is the best between dexamethasone and methylprednisolone, and more importantly, future research should be directed at investigating possible biomarkers of the hyperinflammatory state that could serve as a guide to modulate the dose and duration of GC therapy on an individual basis. In this regard, there is still uncertainty as to why even younger patients without comorbidities sometimes have a more severe course of both SARS-CoV-2-related and unrelated pneumonia. One possible explanation stems from the high variability in the plasma GC levels following administration of the same exogenous dose, which may be related to some degree of glucocorticoid resistance observed in some individuals in terms of CRP reduction and time to clinical improvement [16]. Consequently, it is likely that the GC dose should be titrated according to the level of GC resistance. Some cytokines have been proposed as possible biomarkers of GC resistance, e.g., IL-8, but their role and possible clinical application require further studies in larger populations [14].

Tyrosine Kinase Inhibitor
Janus kinase/signal transducer and activator of transcription proteins (JAK/STAT) is the key signalling pathway shared by the various inflammatory mediators involved in cytokine storms (especially IL-6, INF-γ, (G-CSF), IL-2, and others), resulting in the attraction of macrophages and neutrophils responsible for tissue damage [40].
Based on this notion, treatments with JAK/STAT inhibitors, such as baricitinib or ruxolitinib, selected JAK1/JAK2 inhibitors, have been proposed in COVID pneumonia [41]. In May 2022, the FDA approved the use of baricitinib in hospitalised adults with COVID-19 infection requiring oxygen supplementation, non-invasive mechanical ventilation, or ECMO. This position, derived from previous evidence, has been shared by all major guidelines for the management of severe COVID-19 [42,43].
The first study, called COV-BARRIER, was conducted in 2021 as a randomised, doubleblind, parallel-group, placebo-controlled trial in which hospitalised patients with evidence of SARS-CoV-2 pneumonia were randomly assigned to receive baricitinib or placebo, in addition to standard of care. The study showed a reduction in 28-and 60-day mortality in patients hospitalised in the baricitinib plus standard of care arm; specifically, 39% of patients died in the baricitinib group compared to 58% on placebo-a 19% absolute risk reduction.
The authors pointed out that COV-BARRIER was an initial exploratory study, so they conducted a phase 3 study to further investigate the results, studying the effect of baricitinib in critically ill patients requiring invasive mechanical ventilation (IMV) or extracorporeal membrane oxygenation (ECMO). The results of this phase 3 study confirmed a reduction in all-cause mortality at 28 and 60 days (38.2%) in the baricitinib group compared with placebo [44].
Confirming previous results, the ACTT-1 group conducted a randomised, doubleblind, placebo-controlled trial evaluating remdesivir alone versus remdesivir plus baricitinib in a group of 1033 patients hospitalised with COVID-19 pneumonia, including the subgroup receiving high-flow oxygen (HFO) or non-invasive ventilation (NIV). This study demonstrated reduced recovery time and rapid improvement in clinical status in the baricitinib plus remdesivir arm, especially in patients who required HFO or NIV [45].
Bronte et al. also examined the effect of baricitinib on serum cytokine levels. This was a longitudinal observational study in which 20 patients with SARS-CoV-2 pneumonia were treated with baricitinib plus standard care, showing a decrease in serum levels of IL-6, IL-1β, and TNF-α and an increase in lymphocyte cells, as well as a reduction in the need for oxygen therapy with a progressive increase in the P/F ratio. This study dates back to the early stages of the pandemic, and it is important to mention that none of the included patients received systemic steroid therapy [46].
Finally, the RECOVERY study group enrolled 10,852 inpatients, 8156 of whom received standard care with baricitinib or alone. This is the largest randomised trial that tested the effect of Jak INH in patients hospitalised with SARS-CoV-2 infection and showed a significant reduction in 28-day mortality in the baricitinib lineage (12% in the baricitinib group vs. 14% in the usual care group). However, RECOVERY also shows that the benefit of baricitinib therapy was less than suggested by previous studies [47].
Cao et al., conducted a multicentre, single-blind, randomised, controlled trial of 43 hospitalised patients randomly assigned to receive ruxolitinib or placebo both with standard of care and showed no faster clinical improvement in patients who received ruxolitinib, but lower levels of cytokines in the blood (IL-6, nerve growth factor b, IL-12 (p40), macrophage migration inhibitory factor, MIP-1a, macrophage inflammatory protein 1b (MIP1b), and vascular endothelial growth factor (VEGF)) in the ruxolitinib group [48].
These data were confirmed by the randomised, double-blind, placebo-controlled, phase 3 RUXCOVID trial, which demonstrated no benefit in treatment with ruxolitinib in 432 patients hospitalised with SARS-CoV-2 pneumonia who required oxygen supplementation, randomly assigned to receive ruxolitinib or placebo with the standard of care [49].
The studies available to date have therefore endorsed the use of baricitinib, with the FDA's current indications approving the use of this drug for adult hospitalised patients treated with high-dose oxygen, NIV, or ECMO and systemic steroids. In contrast, however, the EMA has not yet approved the use of this drug in Europe. The above-mentioned RCTs currently ongoing may, in the future, offer therapeutic alternatives to the use of baricitinib [50,51].

Anti-Cytokine Treatment
IL-1 and IL-6 play key roles as mediators of inflammation secondary to SARS-CoV-2 infection. The dysregulation and overactivation of signalling pathways mediated by these cytokines are implicated in the development of severe COVID-related symptoms [52].
Specifically, dendritic cells and mononuclear macrophages, activated by PPRs (pattern recognition receptors), secrete IL-1 and IL-6, which maintain and sustain the acute inflammatory reaction through the activation of innate immunity. IL-6 also interacts with T cells and is responsible for dysregulated T-cell activation in SARS-CoV-2 disease [40].
In light of the scientific evidence from the various clinical trials, in June 2021, the US Food and Drug Administration (FDA) cleared tocilizumab in hospitalised patients (both adults and paediatric) receiving systemic steroids and requiring supplemental oxygen, NIV, or ECMO.
The RECOVERY group enrolled 4116 COVID-19 patients with hypoxia and increased C-reactive protein (CRP) who were randomised to receive tocilizumab or placebo plus standard of care and found that tocilizumab reduced 28-day mortality (31% of patients receiving tocilizumab died at 28 days vs. 38% of patients treated with standard of care) and progression to NIV (35% vs. 42% respectively; hazard ratio 0.84) [55].
At the same time, Rosas et al. enrolled 452 patients with severe COVID-19 pneumonia to receive tocilizumab versus placebo in a 2:1 ratio (COVACTA study) and found no significant differences in mortality between the two groups, but the secondary outcome showed a reduction in ICU and hospital discharge times [56]. The EMPACTA trial of 389 patients hospitalised for severe COVID-19 pneumonia who did not require NIV, randomised to receive tocilizumab or placebo, confirmed that tocilizumab therapy did not change mortality, but reduced the risk of worsening respiratory function and the need for NIV [57].
Somers et al. evaluated the efficacy of tocilizumab with an observational, controlled study focusing on a subgroup of critically ill, mechanically ventilated patients with SARS-CoV-2 pneumonia. This study confirmed a reduction in the mortality in patients treated with tocilizumab. In addition, the incidence of secondary bacterial nosocomial pneumonia was shown to be increased in these patients, but did not affect the final outcome [58].
The REMAP-CAP research group compared tocilizumab or sarilumab versus placebo in 803 critically ill COVID-19 patients in the ICU receiving organ support, enrolled within 24 h. This study confirmed a reduction in mortality among patients treated with IL-6 receptor blockade compared with controls (27% vs. 36%, respectively), associated with a shorter recovery time. In addition, the REMAP-CAP emphasised the importance of identifying patient characteristics that may predict better response to IL-6R inhibitor drugs, such as blood levels of RCP and IL-6, and the importance of early treatment in patients at higher risk of SARS-CoV-2 disease progression [59].
The evidence on the use of sarilumab is still not unambiguous; for example, an Italian open-label study of 56 patients with severe COVID-19 pneumonia resulting in respiratory failure and elevated blood levels of IL-6 found no reduction in mortality in the group of 28 patients treated with sarilumab [60].
Confirming the previous findings, the Sarilumab COVID-19 Global Study Group conducted a multinational, phase 3, randomised, double-blind, placebo-controlled, 60-day study involving 420 patients with COVID-19 who required supplemental oxygen, randomly assigned to receive 200 or 400 mg of sarilumab versus placebo. In this cohort of patients, there was no reduction in mortality following treatment with sarilumab, but it should be considered that a wide range of patients with different baseline characteristics and, thus, wide variability in oxygen requirements and blood levels of IL-6 or CRP were examined. It should be noted that in patients who required non-invasive or invasive ventilation or ECMO, a 9% higher 29-day survival rate was found in the sarilumab-treated group compared with placebo, which might suggest that treatment with this monoclonal antibody becomes meaningful in more severe patients [61].
Regarding siltuximab, at least four clinical trials are currently underway to determine the possible efficacy of this drug in patients with critical illness. In particular, an Italian observational study (SYLVANT) recently completed the enrolment phase of patients with ARDS secondary to COVID-19 pneumonia treated with siltuximab.
The SAVE-MORE trial is a double-blind randomised controlled trial that proposed the use of anakinra in patients with rapidly deteriorating respiratory function. It also proposed a set of parameters that can enable physicians to quickly identify this type of patient so that effective treatment can be set up as soon as possible. Specifically, SAVE-MORE proposed using the level of soluble urokinase plasminogen activator receptor (suPAR) in the blood as an index of the rapid deterioration of respiratory function, and based on the values found, it proposes treatment with anakinra. The importance of this study lies in the fact that it emphasises how early treatment, which anticipates the damage induced by the inflammatory hyperactivation typical of severe forms of SARS-CoV-2, is able to change the course of the disease by significantly reducing the risk of worse clinical outcomes at day 28 [62].
Huet et al. explored the use of anakinra in patients hospitalised with severe SARS-CoV-2 through the prospective Ana-COVID cohort study, which included 99 patients, 52 of whom were treated with anakinra characterised by a reduction in the need for invasive mechanical ventilation in the ICU and mortality [63].
Finally, a systematic review and patient-level meta-analysis by the International Collaborative Group for Anakinra in COVID-19 analysed the vast body of available literature and confirmed that anakinra could be useful in patients with COVID-19, especially in those who show signs of hyperinflammation, e.g., elevated blood PCR values [64].
On the basis of this amount of data, tocilizumab has been confirmed for use in clinical practice to treat severe SARS-CoV-2 pneumonia; anakinra is also suggested for the treatment of COVID-19 infections resulting in the need for supplementary oxygen, with this drug being particularly useful in preventing progression to forms of severe respiratory failure. The data available so far, however, have emphasised that anakinra is only effective, and therefore indicated, if the treated patients have high blood levels of suPAR [4,50,51].

Anti-Complement Therapies: Anti-C5a
The innate immune response to the pathogen leads to the activation of complement pathways (lectin, classical, or alternative pathways), particularly after SARS-CoV-2 infection. The key complement components C3 and C5 are cleaved by C3 and C5 convertases into active C3a and C3b, and C5a and C5b, respectively. C3a and C5b are active anaphylatoxins that mediate the inflammatory response [65]. C5a, in particular, plays a key role in maintaining inflammation by recruiting and activating neutrophils and monocytes. Accordingly, several studies have suggested the use of drugs that can block chain events related to complement activation to modulate the inflammatory response [66,67].
In addition, elevated serum levels of C5a have been found in patients with severe SARS-CoV-2 pneumonia [68].
The PANAMO clinical trial is an exploratory, open-label, randomised trial that tested the benefit and safety of vilobelimab, an IFX-1 monoclonal antibody, a selective blocker of C5a. Initially, a phase 2 study enrolled 30 patients with severe COVID-19 pneumonia, documented by radiologic findings of pulmonary infiltrates, PaO2/FiO2 between 100 mm Hg and 250 mm Hg, or need for invasive or non-invasive ventilation, to be randomly assigned to receive standard therapy alone or more vilobelimab. This study demonstrated the benefits and safety of vilobelimab [69]. Thus, in 2022, the PANAMO trial initiated a multicentre, phase 3, randomised, double-blind, placebo-controlled study that included 368 patients with severe COVID-19 pneumonia undergoing invasive mechanical ventilation, with PaO2/FiO2 ratios of 60-200 mm Hg, who received vilobelimab or placebo. This study showed a reduction in mortality at 28 days (32% in the vilobelimab arm versus 42% in the placebo arm) and at 60 days (35% versus 46%, respectively) in patients receiving vilobelimab, suggesting the use of these drugs as adjunctive therapy in critically ill inpatients [70].
Several clinical trials testing the efficacy of different drugs that block C5 activity are currently underway, notably a Belgian clinical trial testing Zilucoplan ® vs. placebo (NCT04382755) and a French trial testing avdoralimab has completed recruitment (NCT04371367).
Zilucoplan ® and avdoralimab are drugs recently approved for use in clinical practice and therefore were tested later than vilobelimab in SARS-CoV-2 pneumonia. Ongoing studies could provide additional data for the use of these treatments.

Interferon
After viral infection, the innate immune system responds with interferons (IFNs) to limit the severity of inflammation due to its immunomodulatory and antiviral effects.
In 2020, the COVID-19 study group on inhaled interferon beta proposed the use of nebulised interferon beta-1a as adjunctive therapy in hospitalised patients with SARS-CoV-2 infection in a phase 2, randomised, double-blind, placebo-controlled pilot study. This study included 101 patients with mild, moderate, and severe disease and showed an overall higher probability of improvement and faster recovery from SARS-CoV-2 infection in patients receiving nebulised IFN β-1a [71].
The ACTT-3 study group conducted a double-blind, randomised, placebo-controlled trial of 969 patients with COVID-19 pneumonia who had radiographic infiltrates on imaging, SpO2 with room air of 94% or less, or who required supplemental oxygen. Treatment with remdesivir alone versus remdesivir plus interferon beta-1a administered subcutaneously was proposed. This study showed no benefit in the remdesivir plus IFN β-1a branch; moreover, patients who required high-flow oxygen showed a worse outcome when treated with IFN [72].
Several ongoing clinical trials are currently testing the efficacy of inhaled IFN, particularly in hospitalised patients with COVID-19 requiring oxygen therapy in the COV-NI trial (NCT04469491) and in children (NCT05381363).
The current data therefore do not support the use of interferon in COVID-19 pneumonia; ongoing studies could probably identify specific populations that may benefit from interferon therapy.

Anti-GM-CSF
Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a cytokine released by macrophages, lymphocytes, endothelial cells, fibroblasts, and alveolar epithelial cells with immunomodulatory and pro-inflammatory effects. In the lungs, GM-CSF binds GM-CSF receptors located on type II alveolar epithelial cells and macrophages, activating their immune function and supporting surfactant homeostasis [73]. High blood levels of GM-CSF have been found in patients with severe COVID-19 disease. Furthermore, elevated levels of this cytokine appear to be specific for SARS-CoV-2 disease and have not been detected in other viral lung infections, such as influenza. In addition, the measurement of serum levels of IL-6 and GM-CSF has been proposed for the stratification of the severity of COVID-19 infection [74].
The SARPAC clinical trial is a Belgian prospective, randomised, open-label, interventional study investigating the efficacy of sargramostim in patients with acute COVID-19 disease resulting in hypoxic acute respiratory failure, currently under publication (NCT04326920). Bosteels et al. proposed inhaled therapy with sargramostim 125 mcg twice daily for 5 days plus standard of care compared with placebo. In addition, patients who developed ARDS and required mechanical ventilation within the 5-day period switched to sargramostim 125 mcg/m 2 body surface area intravenously once daily until day 5 [78].
There are currently several clinical trials testing the efficacy of intravenous sargramostim in patients with hypoxemia-associated COVID-19 disease (NCT04411680) and inhaled sargramostim to prevent ARDS (NCT04569877) or in mild to moderate SARS-CoV-2 disease to prevent progression to severe forms (NCT04707664; NCT04642950).
Temesgen et al. conducted a case-cohort study of 39 patients hospitalised with COVID-19 pneumonia and a risk factor for disease progression: 12 patients were treated with three doses of lenzilumab 600 mg intravenously, while 27 patients constituted the control group and were only treated with standard care. The lenzilumab group showed faster improvement and greater reduction of inflammatory markers in the blood [77].
Current guidelines do not yet suggest treatment with GM-CSF inhibitors, as there are insufficient data available in the literature to guarantee their efficacy [79].

Current Guidelines
Current WHO guidelines strongly recommend, for severe and critical forms of COVID-19, the use of corticosteroids together with an IL-6 receptor blocker and baricitinib. In addition, the use of ruxolitinib or tofacitinib is suggested only in the case of unavailability of baricitinib or other IL-6 receptors [4].
Currently, the US Food and Drug Administration (FDA) has only approved baricitinib and tocilizumab for patients with severe SARS-CoV-2 pneumonia who require supportive oxygen, non-invasive or invasive mechanical ventilation, or extracorporeal membrane oxygenation (ECMO), as listed in Table 2. Anakinra is licensed for treatment with ruxolitinib or tofacitinib. Anakinra is licensed for the treatment of COVID-19 in hospitalised adults with pneumonia who require supplemental oxygen and who are at risk of progressing to severe respiratory failure and are likely to have elevated plasma-soluble urokinase plasminogen activator receptor (suPAR), according to the FDA's Emergency Use Authorization [50]. Currently, in the European Union, the European Medicines Agency (EMA) has only approved tocilizumab for intravenous infusion for patients with severe COVID-19 already on corticosteroid therapy. The EMA has also approved anakinra with the same indications as the FDA, specifically requiring blood levels of suPAR of at least 6 ng/mL [51].

Discussion
In addition to supportive therapy, glucocorticoids have become the mainstay and standard of care for severe COVID-19. In particular, current guidelines confirm the evidence that GC administration reduces mortality, with effectively manageable side effects in clinical practice, mostly related to the development of hyperglycaemia or hypernatremia. The use of corticosteroids is also particularly manageable in clinical practice compared to the other agents proposed for the treatment of SARS-CoV-2 infections, given physicians' greater experience with these types of drugs. Currently, the guidelines do not recommend a single molecule; treatment can be administered orally or systemically (taking into account patient characteristics), treatment duration varies from at least 5 to 14 days, and dosing depends on the type of molecule used [4,15,43,79]. Further studies should focus on which steroid treatment protocol is the most appropriate and which molecules are best suited to control the enormous inflammatory response that characterises the cytokine storm.
One possible area of investigation could be the efficacy of combination therapy with monoclonal antibodies and steroid therapy. In addition, an update of the European Respiratory Society (ERS) guidelines, published in June 2022, and the current WHO guidelines, updated in January 2023, strongly recommend the use of monoclonal antibodies against IL-6 and JAK inhibitors together with GCs. In contrast, conditional recommendations have been made for the use of IFN-β and of IL-1 receptor antagonist monoclonal antibody [4,43].
The current results already suggest the option of treating the patient according to the disease phenotype, considering the severity of COVID-19 pneumonia at onset, the inflammatory status detectable by measurement of serum inflammation indices, and the level of respiratory support required (e.g., oxygen only, non-invasive or invasive ventilation). Future research will also likely lead to the identification of genomic or transcriptomic patterns that could allow a deeper understanding of the disease and could have both prognostic and therapeutic relevance.
The discovery of these mechanisms could also clarify why, in the pre-vaccine era, the spectrum of disease presentation was so differentiated in patients with similar baseline characteristics.
Currently, several classes and treatment regimens are available for the treatment of patients with severe COVID-19. Several studies, including large numbers, have demonstrated the greater or lesser efficacy of these drugs and have allowed the development of guidelines for their use based on host characteristics. What is critical to understand today is the correct timing of the initiation of these therapies to ensure their efficacy. It will therefore be essential to identify the biological markers that can guide the clinician not only in choosing the appropriate therapy, but also in the correct therapeutic timing. This evidence and future discoveries bring us closer to tailored therapy based on the clinical features, severity of the disease, and its biological and genetic profile.

Conflicts of Interest:
The authors declare no conflict of interest.