GM-CSF in the treatment of COVID-19: a new conductor in the pathogenesis of cytokine storm?

The outbreak of a new strain of the coronavirus family called severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in China in December, 2019, has quickly turned into a pandemic that, as of June 10, 2020, accounts for more than 7 million confirmed cases and 400 000 deaths worldwide. COVID-19 has an extremely variable disease course, ranging from asymptomatic or oligosymptomatic patterns to severe interstitial pneumonia that can be complicated with acute respiratory distress syndrome and even death.¹ The rapid evolution of the global health emergency and the scarcity of specific antiviral treatments have immediately focused research towards the identification of a proper management strategy to prevent mortality due to COVID-19. In this respect, improved knowledge about the interaction between SARS-CoV-2 and the immune system has led to the identification of a massive release of pro-inflammatory mediators linked to an aberrant immune response as the main cause of infection evolution towards life-threatening respiratory impairment.² This aberrant immune response closely resembles a similar reaction observed following the oncological use of chimeric antigen receptor (CAR) T cells, known as cytokine release syndrome. A high concentration of interleukin-6 (IL-6) observed in cytokine release syndrome, suggesting a pivotal role of this cytokine in the pathogenesis of this condition, paved the way for use of IL-6 inhibitors to treat this complication.³ Similarly, the anti-IL-6 agent tocilizumab was the first biological drug used in the treatment of severe COVID-19 and is still being tested in several ongoing randomised trials. However, the first published results on this approach are controversial and, so far, have not confirmed the effectiveness of IL-6 blockade.4, 5 In addition, inhibitors of other pro-inflammatory mediators involved in the development of cytokine release syndrome are being tested as potential treatment targets in COVID-19.⁶

In this scenario, the study published in The Lancet Rheumatology by Giacomo De Luca and colleagues⁷ is an important step forward in understanding the pathways implicated in the pathogenesis of life-threatening SARS-CoV-2 infection. The Article reports a prospective analysis aimed at evaluating the effect of mavrilimumab, a granulocyte–macrophage colony-stimulating factor (GM-CSF) receptor inhibitor, in the treatment of non-mechanically ventilated patients with COVID-19 pneumonia and systemic hyperinflammation. The authors observed that, by day 28 of follow-up, all 13 patients who received mavrilimumab showed clinical improvement compared with 17 (65%) of 26 patients in the control group (p=0·030), who received standard care provided at the hospital at the time of the study, including hydroxychloroquine, azithromycin, and lopinavir–ritonavir. Furthermore, no patients who received mavrilimumab died, but seven (27%) patients in the control group died by day 28. The study is certainly affected by some potential limitations. The small sample size, the absence of randomisation, and the short follow-up period might reduce the fully generalisability of observed results. However, the inclusion of an overall well matched control group represents a strength compared with other studies evaluating biological drugs for the same indication. In particular, the results reported by De Luca and colleagues on mavrilimumab seem to be overall more favourable than those observed in similarly designed studies exploring tocilizumab4, 5 or the IL-1 inhibitor anakinra.⁸ This evidence, if confirmed by larger randomised studies, might lead to a reappraisal of the role of the components involved in the development of cytokine release syndrome. Certainly, GM-CSF has very complex immunological activity, ranging from the well known haematopoietic effect to the more recently demonstrated pro-inflammatory role, which has made it a potential target for the treatment of immune-mediated diseases such as rheumatoid arthritis, spondyloarthritis, and giant-cell arteritis. Undoubtedly, within the inflammatory cascade, this effect can be placed further upstream compared with other cytokines such as IL-1, tumor necrosis factor, and IL-6.⁹ For an example, a Chinese study reported that atypical pathogenic T helper 1-cells expressing GM-CSF have been detected only in patients with more severe COVID-19 compared with both patients with less symptomatic patterns and healthy controls.¹⁰ These findings seem to confirm a crucial and upstream role of GM-CSF in the pathogenesis of SARS-CoV-2-related hyperinflammation. Another point to consider is the favourable safety profile of mavrilimumab observed both in the study by De Luca and colleagues and in the entire drug development programme for rheumatoid arthritis, for which the reported incidence of serious infections was negligible.⁹ In particular, some concerns could arise from the involvement of GM-CSF in the clearance of the alveolar surfactant by resident macrophages and the development of pulmonary proteinosis in cases of congenital deficit of this mediator. However, no such complication has ever been reported throughout the development programme of all GM-CSF inhibitors.⁹

In conclusion, GM-CSF receptor blockade is certainly a potential option for the treatment of more severe subsets of COVID-19, and more extensive studies are warranted to confirm the role of GM-CSF in the pathogenesis of cytokine release syndrome.

© 2020 Shutterstock