Blocking inflammation on the way: Rationale for CXCR2 antagonists for the treatment of COVID-19

The immune response upon infection exhibits characteristics that may explain the propensity toward a hyperinflammatory phenotype. SARS-CoV-1 and MERS-CoV were shown to circumvent immune recognition by several mechanisms including a potent inhibition of the IFN-inducing signaling pathways. Very recently, Blanco-Melo et al. (2020) demonstrated the same for SARS-CoV-2 using infection models in cell culture, in a ferret animal model, and in human COVID-19 serum samples. The inhibition of the IFN-mediated antiviral immune response, which would effectively block viral replication, leads to an unbalanced immune response with excessive secretion of pro-inflammatory cytokines and leukocyte attracting chemokines (Blanco-Melo et al., 2020), both of which may be targeted by therapeutic intervention.

Elevated serum concentrations of pro-inflammatory cytokines and chemokines, especially IL-1β, IL-6, IL-8, TNF, G-CSF, and GM-CSF, have been described and were shown to correlate with disease severity in COVID-19 patients. The exacerbated cytokine release, also known as “cytokine storm,” has drawn much attention and is being targeted in many clinical studies. Several clinical studies using the IL-6 receptor–blocking antibodies tocilizumab, sarilumab, or the IL-6–blocking antibody siltuximab are ongoing. IL-6 receptor inhibition is effectively used to relieve CAR-T cell therapy-induced cytokine storm, and the first anecdotal reports in COVID-19 treatment raise hope (Vabret et al., 2020). The list of therapeutic interventions targeting the overactive innate immune system is growing, with several studies investigating the IL-1 receptor antagonist anakinra or the JAK1/2 inhibitor ruxolitinib (Vabret et al., 2020).

The described mediators are increasingly produced as a result of massive infiltration of innate immune cells and their overactivation. Another approach may be to inhibit the migration of these pro-inflammatory immune cells to the site of infection in the first place to interrupt the self-propagating hyperinflammation and resulting tissue destruction. Inflammatory monocyte-derived macrophages are found amplified in the bronchoalveolar fluid of severe cases of COVID-19 (Liao et al., 2020) and may play an important role in the self-propagating inflammation by recruiting additional monocytes and neutrophils through high expression levels of the chemokines CCL2, CCL3, CCL5, CXCL8 (IL-8), CXCL9, CXCL10, and CXCL11 (Liao et al., 2020). First clinical studies using the CCR5 antagonist leronlimab (NCT04343651, NCT04347239) and maraviroc (NCT04441385, NCT04435522), originally developed against HIV infection, for the treatment of COVID-19 are ongoing or planned to block the migration of inflammatory monocytes to the site of infection. Besides monocytes, several findings hint toward an active involvement of neutrophils in disease pathogenesis. Neutrophils are known to play an important role in the establishment of lung injury and ARDS (Németh et al., 2020). The neutrophil effector functions—release of antimicrobial and inflammatory substances such as ROS, extracellular proteases like elastase and metalloproteinases, cationic peptides, and neutrophil extracellular traps (NETs)—act potently against extracellular pathogens like bacteria but may be less effective against intracellular pathogens like viruses. An errant overreaction may cause severe collateral damage in the lung: endothelial injury is followed by alveolar edema, NET-mediated inflammasome activation, platelet aggregation, and formation of microthrombi that are increasingly observed in COVID-19 patients (Bikdeli et al., 2020). The resulting diffuse alveolar damage with apparent macrophage and neutrophil infiltrates is indeed found in coronavirus-induced lung pathology (Barnes et al., 2020). Elevated concentration of markers of NET formation, such as cell-free DNA and myeloperoxidase-DNA, are detected in patients and sera from these patients are able to induce formation of NETs by neutrophils form healthy donors (Zuo et al., 2020,Preprint). In a recent Perspective, Barnes et al. (2020) suggest that aberrant NET formation may contribute to organ damage, thrombosis, and cytokine secretion and discuss NETs as a potential drug target. Further evidence for neutrophil involvement is given by the fact that neutrophil counts are increased in severe versus mild cases and the neutrophil-to-lymphocyte ratio is a predictive marker of mortality (Vabret et al., 2020).

Neutrophils are mainly recruited via ligands of the CXC chemokine receptors CXCR1 and CXCR2. A plethora of ligands of these receptors have been described, namely CXCL1, CXCL2, CXCL3, CXCL5, CXCL6, CXCL7, and CXCL8. CXCL8 (IL-8), a major neutrophil-attracting chemokine, is produced by lung epithelial cells upon infection and found to be highly expressed in patients with severe COVID-19 infection. Its serum levels correlate with disease severity (Vabret et al., 2020). One phase II study using the anti-CXCL8 (IL-8) drug BMS-986253 in cancer patients with COVID-19 was recently posted on ClinicalTrials.gov (NCT04347226). However, due to the high redundancy of ligands for CXCR1 and CXCR2, many of which are overexpressed in COVID-19 patients, blocking the receptors itself may be the more promising strategy (illustrated in Fig. 1).

Genetic deletion or therapeutic inhibition of CXCR2 has been reported to improve the course of many inflammatory lung disease models in mice. It was shown in an influenza-induced lung injury mouse model that a CXCR2 antagonist in combination with an antiviral treatment significantly improved clinical score and body weight change, decreased neutrophil count in BAL and reduced alveolar injury compared with antiviral treatment alone (Washburn et al., 2019). Importantly, another preclinical study found that CXCR2 deficiency prevents neutrophil accumulation in the lung in a murine influenza model without compromising viral clearance (Wareing et al., 2007). Several small-molecule antagonists blocking CXCR2 are under clinical investigation for the treatment of inflammatory diseases and cancer. The most developed ones, AZD5069 (AstraZeneca), navirixin (Merck Sharpe & Dohme), danixirin (GlaxoSmithKline), and the combined CXCR1/2 inhibitor reparixin (Dompé Farmaceutici S.p.A.) have completed phase I and II studies. Importantly, they displayed beneficial safety profiles without increasing risk of infection (Roberts et al., 2019; Lazaar et al., 2011; Hastrup et al., 2015; Miller et al., 2015). Studies with ozone- or LPS-induced airway inflammation in human volunteers showed that treatment with CXCR2 antagonists reduced neutrophil counts in the lungs by 50–80% (Lazaar et al., 2011; Leaker et al., 2013). For danirixin, it was shown in a phase IIa study in patients with acute, uncomplicated influenza that the treatment did not impede viral clearance (Roberts et al., 2019), again arguing that CXCR2 antagonism is a safe therapeutic option for virally induced lung injury. In addition to its effect on neutrophil migration, CXCR2 blockade also reduces activation of neutrophils and may therefore exert beneficial effects on a systemic level and reduce NET-induced thrombosis.

Thus, immunopathological evidence of neutrophil involvement in COVID-19, effectiveness of CXCR2 inhibition in preclinical models, and beneficial safety profile in patients with respiratory disease support the idea of applying CXCR2 antagonists in clinical trials for the treatment of severe COVID-19. CXCR2 ligand concentrations, e.g., CXCL1, CXCL2, CXCL5, and CXCL8, in sputum or bronchoalveolar lavage fluid of patients may be a useful biomarker for predicting an upcoming neutrophil infiltration with aggravation of symptoms in order to choose the ideal treatment start.

As of July 14, a search of the 2,553 studies for COVID-19 on https://www.clinicaltrials.gov disclosed very few studies targeting recruitment of myeloid cells (Table 1). None of the studies aim to inhibit neutrophil migration and activation via blockade of CXCR1 and CXCR2. This comment will hopefully spark discussions about using inhibitors of chemokine/chemokine receptor pathways to block excessive infiltration of neutrophils to interrupt the self-reinforcing hyperinflammation in severe cases of COVID-19 infection. Based on the current knowledge of immunopathology of COVID-19 and ARDS, there is strong evidence to investigate the usage of CXCR2 antagonists in the treatment of severe COVID-19.

This work was supported by grants from Else Kröner-Fresenius-Stiftung (2017_A50) and the Deutsche Forschungsgemeinschaft (Projektnummer 442265435; 329628492 – SFB 1321).