Could treatment with immunomodulatory agents targeting IL-1, IL-6, or JAK signalling improve outcomes in patients with severe influenza pneumonia? A systematic and narrative review

Roles for IL-6, JAK signalling and IL-1 in influenza pneumonia

IL-6 and JAK signalling. IL-6 is a pleiotropic cytokine, released by macrophages, monocytes and infected epithelial cells, that drives much of the hyperinflammatory response and symptoms seen in cytokine release syndromes. The IL-6 signalling pathway is activated in response to influenza infection²⁴ within which the JAK/STAT signalling cascade plays an important role²⁵). IL-6 concentrations are elevated in patients with severe influenza (and severe COVID-19)^15–17 and correlate with symptoms and signs of influenza virus infection and viral shedding^26,27. High IL-6 concentrations were associated with disease severity and mortality²⁸ and inversely associated with arterial oxygen levels in hospitalized influenza patients²⁹. However, it remains unclear whether elevated IL-6 is a marker of disease progression or whether it contributes to the pathogenesis of severe disease.

The role of IL-6 has been studied in pre-clinical models but with conflicting results. In an influenza mouse model, IL-6 was found to drive muscle dysfunction, which was attenuated by treatment with the IL-6RA tocilizumab³⁰. Another study demonstrated an IL-6 associated inflammatory response in a lethal A(H1N1)pdm influenza mouse model. However, there was no significant difference in viral load, pathology, weight loss or survival between IL-6 knockout mice and wild-type mice²⁸. Because the sensing of influenza viruses by TLR3 primarily regulates the initiation of the proinflammatory response, TLR3 knockout mice have been studied. In comparison with wild-type mice, mice deficient in TLR3 had a reduced inflammatory response (including reduced IL-6), and an increased lung viral load, but survival was increased in the TLR3 knockout mice³¹. In other mice models, IL-6 has been found to be protective^32,33.

IL-6 signalling occurs via JAK/STAT pathways, including JAK 1/2 which are also involved in the intracellular signalling of other proinflammatory cytokines. Inhibition of JAK1/2 reduces the level of IL-6 and other proinflammatory cytokines³⁴ and JAK1/2 inhibitors are effective treatments in COVID-19^9–12. In addition, JAK1/JAK2 signalling may have a role in influenza viral replication³⁵ where their deficiency reduces replication³⁶. JAK1/JAK2 have also been identified as potential anti-viral drug targets³⁶.

IL-1. IL-1 is one of the major cytokines involved in the hyperinflammatory response to influenza virus infection^15,37. The influenza M2 protein activates the NOD-like receptor protein 3 (NLRP3) inflammasome³⁸ resulting in the secretion of IL-1β among other inflammatory cytokines. Severe influenza can lead to an immune response similar to macrophage activation syndrome (MAS). MAS is an inflammasome- and IL-1-mediated process resulting in cytophagocytosis, profound inflammation, and multisystem organ dysfunction³⁹. It is a life-threatening complication of many diseases including infection³⁹. In acute respiratory distress syndrome (ARDS) secondary to influenza there is evidence of cytophagocytosis⁴⁰. Features of MAS include sustained fever, hyperferritinemia, high IL-18, pancytopenia, fibrinolytic consumptive coagulopathy, and liver dysfunction.

Influenza is associated with secondary Staphylococcus aureus pneumonia and invasive aspergillosis. The combination of these secondary infections and MAS is consistent with some clinical features of chronic granulomatous disease (CGD), a rare severe immunodeficiency caused by a mutation in one of the genes encoding the NADPH-oxidase complex proteins^41–43. In pre-clinical studies, influenza virus infection has been shown to suppress the NADPH-oxidase complex resulting in increased susceptibility to S. aureus infection⁴⁴. The increased risk of these infections in severe influenza could represent a transient CGD-like phenotype.

Potential treatment with immune modulators targeting IL-1, IL-6 and JAK signalling in influenza

IL-6 receptor antagonists. Tocilizumab and sarilumab are monoclonal antibodies that target both membrane-bound and soluble IL-6 receptors (‘IL-6 RAs’). Tocilizumab was first approved in 2010 by the Food and Drug Administration (FDA) for the treatment of rheumatoid arthritis. It has received additional approval for giant cell arteritis, systemic and juvenile forms of idiopathic arthritis, and in 2017, for severe or life-threatening CAR T-cell induced cytokine release syndrome (CRS). Recently, IL-6RAs have shown benefit in hospitalized patients with COVID-19. The Randomised, Embedded, Multi-factorial, Adaptive Platform Trial for Community-Acquired Pneumonia (REMAP-CAP; NCT02735707) demonstrated the efficacy of tocilizumab and sarilumab⁶, in improving outcomes of critically ill COVID-19 patients, including survival and organ support-free days. In the Randomised Evaluation of COVID-19 Therapy (RECOVERY, NCT04381936) trial, tocilizumab reduced mortality in hospitalized COVID-19 patients⁷. A WHO meta-analysis concluded that IL-6RAs reduce 28-day all-cause mortality in severe COVID-19 when given in addition to corticosteroids⁸.

Tocilizumab has a strong safety profile in ambulatory patients with rheumatoid arthritis and other inflammatory diseases, with increased risk of infection being the most common serious adverse events (five per 100 patient-years (PY) compared to four per 100 PY for other disease-modifying agents, https://www.actemrahcp.com/ra/clinical-study-safety).

A small retrospective observational study compared patients with juvenile idiopathic arthritis (JIA) who were being treated with tocilizumab and developed influenza virus infection (10 patients, mean age 14 years) to similar patients on non-biologic treatments⁴⁵. Tocilizumab treatment did not increase the risk of severe disease with influenza when compared to controls. In contrast, patients receiving tocilizumab had less fever, a shorter duration of fever and lower c-reactive protein (CRP) levels. Neither group experienced severe complications, such as pneumonia. The relevance of these data is limited due to the low patient numbers, the study design and the long-term JIA tocilizumab treatment compared with acute treatment for severe influenza pneumonia but suggests that IL-6RA may moderate influenza severity.

The REMAP-CAP⁶ and RECOVERY⁷ trials did not observe increased rates of secondary infections in patients with COVID-19 receiving treatment with IL-6RAs. In the WHO meta-analysis, IL-6RAs increased survival in severe COVID-19, without significantly increasing secondary bacterial infections (21.9% of patients treated with IL-6RAs vs 17.6% treated with usual care or placebo, OR 0.99; 95% CI, 0.85–1.16)⁸.

IL-1 receptor antagonists. Anakinra is a recombinant IL-1 receptor antagonist (IL-1RA) that competitively inhibits IL-1α and IL-1β binding to the IL-1 type I receptor. Despite a lack of randomised trial data, anakinra has been used to treat both paediatric⁴⁶ and adult⁴⁷ MAS. The SAVE-MORE multicentre placebo-controlled trial assessed treatment with anakinra in COVID-19 patients (n=594, 85.9% of patients also receiving steroids) who were at risk of progressing to respiratory failure (as identified by plasma soluble urokinase plasminogen activator receptor [suPAR] levels ≥6 ng/ml)¹³. Treatment with anakinra improved scores on the 11-point World Health Organization Clinical Progression Scale. 50.4% (204/405) of patients receiving anakinra had fully recovered with no viral RNA detected at day 28 compared to 26.5% (50/189) of control patients (placebo). In the same study, anakinra treatment also reduced 28-day mortality¹³.

However, other studies have found no benefit of anakinra in COVID-19. In REMAP-CAP, there was no benefit amongst critically ill patients¹⁴. The ANACONDA trial (NCT04364009) was suspended (after n=71) due to concern for excess mortality in the anakinra arm⁴⁸. Similarly, no benefit was observed in the CAN-COVID trial of another anti-IL-1β monoclonal antibody canakinumab (n=454 hospitalized patients)⁴⁹. These conflicting results may be due to differences in patient cohorts, disease severities, patterns of hyperinflammation and different trials. Measuring suPAR levels, as in the SAVE-MORE trial, might select patients who are more likely to benefit from IL-1 inhibition, or canakinumab may be less effective (inhibits only IL-1β not IL-1α). Alternatively, there could be a differential treatment effect. It has been suggested that the lack of efficacy in critically ill patients may be explained by higher levels of anti-IL1RA antibodies^50,51. In two large, randomised trials (n=893, n=696) of critically ill patients with sepsis and septic shock anakinra did not reduce overall all-cause mortality^52–54. However, a post-hoc analysis found improved survival in a subgroup of patients with features of MAS (ferritin elevation >2,000 ng/ml, coagulopathy, and liver enzyme elevations)⁵⁵, although this result is only hypothesis generating it does suggest that a subset of patients may benefit from IL-1 inhibition. Anakinra reduced inflammation and protected CGD mice from invasive aspergillosis and was used in a small number of CGD cases without increased risk of severe infection⁵⁶. To date, IL-1 receptor antagonism has not been evaluated in influenza virus pneumonia.

Anakinra has an established safety profile in ambulatory patients with rheumatic and inflammatory diseases. In rheumatoid arthritis (RA) trials, the incidence of serious adverse events (SAEs) with anakinra (100 mg/day) was comparable to placebo while the incidence of severe infection was higher (1.8% vs. 0.7%) and neutropenia occurred more frequently (https://www.ema.europa.eu/en/medicines/human/EPAR/kineret. Importantly, anakinra did not increase the risk of adverse events (AEs) or SAEs in patients with septic shock when compared to placebo^52–54.

JAK/STAT signalling inhibitors. Baricitinib is an oral selective JAK1/JAK2 inhibitor. It has been widely used for autoimmune and atopic diseases. Baricitinib inhibits the intracellular signaling of cytokines known to be elevated in severe influenza, particularly IL-6. In COVID-19 infection, baricitinib has been shown to rapidly reduce the levels of proinflammatory cytokines^18,32, including IL-1, IL-6, and MCP-1, all of which are elevated in severe influenza. Baricitinib is an effective treatment for COVID-19 hospitalized patients. In the RECOVERY trial, treatment with baricitinib in comparison to standard care alone (n=8156, 95% receiving corticosteroids, 23% receiving tocilizumab), baricitinib reduced mortality at 28 days (RR 0·87; 95% CI 0·77−0·99; p=0·028)¹². In the placebo controlled phase III Adaptive COVID-19 Treatment Trial 2 (ACTT-2) (n=1033, patients receiving corticosteroids excluded), treatment with baricitinib and remdesivir reduced the time to recovery in comparison to remdesivir alone (rate ratio 1.16, 95% CI 1.10–1.32, p=0.03) but did not reduce 28-day mortality (5.1% vs 7.8%, hazard ratio 0.65, 95% CI 0.39–1.09)⁹. The ACTT-4 trial assessed baricitinib plus remdesivir versus dexamethasone plus remdesivir in hospitalised patients receiving supplemental oxygen (n=1010)⁵⁷ in a trial conducted before dexamethasone was standard of care for hospitalized adults with COVID-19 requiring supplemental oxygen. Mechanical-ventilation free survival was comparable but baricitinib was associated with reduced AEs and SAEs than dexamethasone. In the COV-BARRIER study, a phase III randomized clinical trial of baricitinib in hospitalized COVID-19 patients not requiring mechanical ventilation (n=1525, 79.3% receiving corticosteroids), there was no difference in disease progression (odds ratio 0·85, 95% CI 0·67–1·08, p=0·18)¹⁰. However, 28-day all-cause mortality was significantly reduced with baricitinib (8% vs 13%, hazard ratio 0.57, 95% CI 0.41–0.78, p=0.0018). Subsequently, COV-BARRIER included critically ill patients requiring mechanical ventilation (86% receiving steroids) and all-cause mortality was significantly lower with baricitinib vs placebo (39.2% vs 58.0%, hazard ratio 0·54, 95% CI 0·31–0·96, p=0·030)¹¹. An updated meta-analysis including the RECOVERY trial (ACTT-4 not yet included) found treatment with baricitinib or another JAK inhibitor was associated with a 20% proportional reduction in mortality in hospitalized COVID-19 patients¹². Baricitinib has not been tested in patients with influenza virus pneumonia. Baricitinib may also have anti-viral effects in influenza. Inhibition of JAK2 has been shown to reduce influenza viral replication in vitro³⁵. JAK1 and JAK2 have been identified in genome-wide screens as potential drug targets for inhibiting influenza virus replication³⁶.

Baricitinib has a good safety profile in long-term patient use with some increased risk of infection identified in RA patients⁵⁸, particularly with herpes zoster⁵⁹. The relevance of such long-term effects in acute treatment for severe influenza pneumonia is unclear. The safety of baricitinib has been demonstrated in hospitalized and critically ill COVID-19 patients. ACTT-2 (n=1033) and ACTT-4 (n=1010) found fewer AEs and SAEs and ACTT-2 also found fewer new infections in the baricitinib group^9,57, with no increased risks found in RECOVERY (n=8156)¹², COV-BARRIER hospitalized (n=1525)¹⁰ and critically ill (n=101)¹¹. There was no difference in rates of venous thromboembolism (reported previously for long-term baricitinib use in RA⁵⁹) in any of the studies^9–12,57.

Systematic search yields no trial evidence targeting IL-1, IL-6, or JAK signalling in severe influenza

A summary of the literature search is shown in the PRISMA flowchart (Figure 1). Five-thousand four-hundred and nine (n=5409) records were screened for eligibility after removing duplicates. Following level I screening, five articles met full eligibility for level II full text screening. Of these, no articles were suitable for inclusion (no RCTs were found testing IL-1, IL-6 or JAK signalling inhibitors in severe influenza).

Figure 1. Study flow diagram.

The most closely relevant studies included the aforementioned study of tocilizumab and influenza severity in JIA patients which is an observational study with matched controls⁴⁵, a single patient case report⁶⁰ and some studies of Chinese Traditional Medicine that may target these immune pathways but the exact mechanisms of action are unclear⁶¹.

The need for randomised controlled trials

Severe influenza pneumonia is a significant public health burden and cause of morbidity and mortality worldwide. Severe influenza often leads to hyperinflammation involving the release of pro-inflammatory cytokines, particularly, IL-1 and IL-6, and signalling through JAK1/JAK2. Treatment with immune modulatory agents, including the IL-6RA tocilizumab and the JAK1/JAK2 inhibitor baricitinib, are beneficial in COVID-19 which shares common features of hyperinflammation with severe influenza. Whether the IL-1RA anakinra is beneficial in COVID-19 or influenza remains unclear. Furthermore, all agents have an established safety profile with both long-term use and in randomised trials for severe COVID-19.

Our extensive systematic review (2022) found no trial evidence available and no trials currently registered (although an influenza immune modulation domain is planned in REMAP-CAP) highlighting an evidence gap. Well-controlled studies are required as immune modulation could be beneficial, ineffective, or even cause harm in severe influenza pneumonia. The presence or absence of bacterial co-infection (infection occurring concomitantly), secondary bacterial infection (occurring after initial influenza infection/hospitalisation/treatment, including ventilator-associated pneumonia VAP) may affect treatment response. Thus, randomised trials in severe influenza pneumonia should assess for a differential treatment effect when bacterial co-infection is present. This may require trial stratification depending on the presence or absence of such infections. Additional evidence gaps include the potential for treatment interactions between immune modulation and other interventions, such as antivirals and corticosteroids, the effect of immune modulation on viral shedding, and the effect of acute immune modulation on long-term outcomes such as Major Adverse Cardiovascular Events (MACE) potentially related to the acute inflammatory response.