The progression of COVID-19 varies significantly among individuals; the severity and mortality rate are correlated with obesity, pre-existing type 2 diabetes and hypertension. However, not all patients from highrisk groups are characterized by severe passing of disease. Finding precise and rapid prognostic indicators is one of the main aims of our study. The study used plasma samples from 103 patients and volunteers. A retrospective analysis was performed on 93 COVID-19 patients classified by the severity of the disease as mild (n = 14), moderate (n = 42), and severe (n = 37). The distribution into groups was conducted after the completion of treatment in patients. The severity of COVID-19 was retrospectively assessed at the time of hospital discharge. Upon admission, all patients presented with similar symptoms of the disease. Glucocorticosteroids were not administered during the pre-hospital treatment phase. Demographic data and parametric indicators were registered. Clinical analysis and quantification of extracellular DNA (cfDNA) were performed and the levels of NETosis, the concentration of human myeloperoxidase (MPO) and human neutrophil elastase in blood plasma were measured. Statistical analyses were provided in the R environment (www.R-project.org, V.4.0). Our data illustrates that the severity of coronavirus disease among subjects largely correlates with the initial cell-free DNA in plasma and NETs formation activity. The plasma cfDNA levels in the early period of the disease were statistically significantly different in the subsequent course of the disease in patients with diabetes. The cfDNA value of 4297 ng/ml and higher corresponded to a more severe passing of disease in patients with COVID-19 from a high-risk group with 82% sensitivity and 72% specificity (AUC0.856, 95% CI 0.778-0.935, P < 0.001). The level of elastase in the blood plasma of patients with severe COVID-19 shows a statistically significant difference from the level of mild and moderate patients (P < 0.001), as well as compared with the level of elastase in healthy donors. In addition, the data demonstrate statistically significant differences in MPO levels between all groups of patients with COVID-19. We found an association between circulating NET formation markers at the disease’s initial stage and clinical outcome. This demonstrates the potential importance of assessing plasma levels of cfDNA for clinical decision-making in the early stages of the disease.