ABSTRACT
Coronavirus disease 2019 (COVID-19) is characterized by immune activation in response to viral spread, in severe cases leading to the development of cytokine storm syndrome (CSS) and increased mortality. Despite its importance in prognosis, the pathophysiological mechanisms of CSS in COVID-19 remain to be defined. Towards this goal, we analyzed cytokine profiles and their interrelation in regard to anti-cytokine treatment with tocilizumab in 98 hospitalized patients with COVID-19. We performed a multiplex measurement of 41 circulating cytokines in the plasma of patients on admission and 3-5 days after, during the follow-up. Then we analyzed the patient groups separated in two ways: according to the clusterization of their blood cytokines and based on the administration of tocilizumab therapy. Patients with and without CSS formed distinct clusters according to their cytokine concentration changes. However, the tocilizumab therapy, administered based on the standard clinical and laboratory criteria, did not fully correspond to those clusters of CSS. Furthermore, among all cytokines, IL-6, IL-1RA, IL-10, and G-CSF demonstrated the most prominent differences between patients with and without clinical endpoints, while only IL-1RA was prognostically significant in both groups of patients with and without tocilizumab therapy, decreasing in the former and increasing in the latter during the follow-up period. Thus, CSS in COVID-19, characterized by a correlated release of multiple cytokines, does not fully correspond to the standard parameters of disease severity. Analysis of the cytokine signature, including the IL-1RA level in addition to standard clinical and laboratory parameters may be useful to define the onset of a cytokine storm in COVID-19 as well as the indications for anti-cytokine therapy.