COVID-19 inflammatory profiles: distinct role for GM-CSF

Expert Name
Prof. Peter Openshaw
Expert Affiliation
National Heart and Lung Institute, Imperial College London, UK

Our study in brief:

We performed a very large collaborative study within the International Severe Acute Respiratory and emerging Infection Consortium (ISARIC) in over 200 different hospitals in the UK. The aim was to identify inflammatory markers in blood which increased within the early stages of COVID-19 in those who went on to become critically ill. We just recently published this data in Science Immunology [1].

One marker in particular, the cytokine granulocyte macrophage colony stimulating factor (GM-CSF), was found at higher levels in those SARS-CoV-2 infected individuals who later progressed to severe disease, compared with healthy controls, or those with influenza. GM-CSF levels were found to be almost 10 times higher in those who died from COVID-19. Increased GM-CSF could help to identify those at risk of developing severe COVID-19 and provide a target for new treatments which could modify their course of disease.

The remarkable thing about this study was how quickly we were able to launch it. We had already put in place a lot of the protocols that we needed in order to do this study. Over 2,500 people within the national health service were responsible for contributing data on the patients and also taking lots and lots of samples. We collected serial samples from over 2,000 patients. It is, in global terms, one of the biggest studies that involves measuring cytokine and inflammatory mediators in patients with COVID who have been hospitalized. Furthermore, these data are from a prospective study, and we collected patients right from the start of the first wave in the UK. To date, over 200,000 patients have been enrolled within the study as a whole.

Specific signals in the immune system responsible for response to COVID-19 infection

We published the analysis on 32 different soluble cytokine mediators in the peripheral blood, and we were able to do this in relation to severity because there were so many patients in the study. We stratified the patients according to greatest severity; we had different levels of severity from mild disease (seen mostly in healthcare workers) to those who required mechanical ventilation, and those who very sadly died, really encompassing the full spectrum of disease. Essentially, most of the mediators that we measured showed an upward trend with severity.

That upward trend particularly applied to mediators like interleukin-6 (IL-6), which was reassuring, since IL-6 is already been robustly associated with disease. One cytokine which came out very strongly from the study was that GM-CSF was very strongly associated with severity, specifically for COVID-19, as compared with influenza (Figure). When we were able to associate all the mediators in a sort of network of connectedness, which was one of the very powerful analyses that were done, we can show that there are distinct hubs for interleukin-6 and GM-CSF, which may gave us the hope, that in addition to inhibiting IL-6, which has now been proven to be effective in a subgroup of patients with COVID-19, inhibiting GM-CSF might also be beneficial. That is the subject of ongoing trials; one such study, Otilimab in Severe COVID-19 Related Disease (OSCAR; NCT04376684), is on-going.  We are waiting for the results of those studies.

Macrophage-monocyte activation

The virus itself is not an extremely cytolytic one. It is often relatively mild in the earliest stages of infection. The distinguishing characteristic that really is emerging as attributable to local damage in the lung, is one of macrophage-monocyte activation.

One other thing which comes out from our cytokine mediator studies, is that we measured excessive markers of endothelial inflammation, including coagulation and endotheleitis. This might explain some of the other manifestations of disease, such as thrombosis within the small and large vessels. In particular, clotting in the peripheral veins might lead to embolism, but also affecting the pulmonary arterial supply, which, like the venous circulation, has deoxygenated blood and a relatively low pressure. There seems to be some endotheleitis in the lung as well, which may contribute to closing-off of vessels, with consequent mismatching between ventilation and perfusion within the lung, leading to hypoxia. Thus, the lung damage is in part due to the activation of macrophage-monocytes within the lung, but also is associated with this imbalance of  ventilation/perfusion, which is a consequence of the endothelial inflammation and therefore thrombosis, in these smaller-medium vessels.

Association between type 2 inflammation, such as seen in allergic asthma cases?

As an adult chest physician, I have always been very interested in Th2 disease, or type 2 inflammation. For COVID-19, asthma does not appear to be a significant risk factor for hospital admission, but there are underlying characteristics that can be parallel to type 2 inflammation. Interestingly, in our study, IL-4 levels were lower in the moderate/non-severe disease outcome groups relative to both control groups and severe disease groups, indicating that suppression of normal type-II cytokines levels may be associated with milder COVID-19, and that this is lost in severe disease. These data partially recapitulate the association of type 2 mediators with COVID-19 severity, though levels are typically low and require careful interpretation relative to control samples.

In terms of the cytokine mediators, it is clear that, initially, there is an early strong interferon-related antiviral response. In those individuals who are capable of developing that response, that disease can sometimes be aborted or truncated, and it then evolves over time through various other forms of inflammation. The macrophage-monocyte inflammation is responsible for the more severe disease in COVID-19, rather than being a Th2 type disease, or an exacerbation of prior inflammation caused by asthma. Nevertheless, the potential parallels with type-2 inflammation and association with disease outcomes remains intriguing and warrants further research.


  1. Thwaites RS, Sanchez Sevilla Uruchurtu A, Siggins MK, Liew F, Russell CD, Moore SC, Fairfield C, Carter E, Abrams S, Short CE, Thaventhiran T, Bergstrom E, Gardener Z, Ascough S, Chiu C, Docherty AB, Hunt D, Crow YJ, Solomon T, Taylor GP, Turtle L, Harrison EM, Dunning J, Semple MG, Baillie JK, Openshaw PJ; ISARIC4C investigators. Inflammatory profiles across the spectrum of disease reveal a distinct role for GM-CSF in severe COVID-19. Sci Immunol. 2021 Mar 10;6(57):eabg9873.

Figure. GM-CSF and IL-1α were elevated in fatal COVID-19 relative to influenza. Levels of IL-6, GM-CSF, IL-1α, IL-1β, Thrombomodulin, and vWF-A2 in plasma samples from patients with fatal influenza (n=20) or COVID-19 (n=69).