Longitudinal Cytokine Expression Profiles Indicate COVID-19 Disease Trajectory

Young ill woman holding hand on forehead, checking temperature, resting, lying on the couch with a cozy blanket. Using purple face mask to prevent other people from getting infected.
Study authors assessed SARS-CoV-2 antibody responses and identified cytokine signatures using hierarchical clustering.

Researchers identified 3 distinct cytokine signatures which are associated with differing COVID-19 trajectories. These findings were published in The Journal of Infectious Diseases.

A subset of patients (N=19) enrolled in the remdesivir clinical trial (ACTT-1) for COVID-19 were recruited for this study. Patient serum was assessed for serology and cytokines.

At baseline, 12 patients were seropositive (50% neutralizing titer [NT50], >80) and 7 were seronegative (NT50,<80). Posttreatment, all patients were seropositive.

Low neutralization activity was observed among patients who were negative for spike receptor-binding domain (RBD) with or without nucleocapsid RBD-specific immunoglobulin (Ig)G. NT50 was highly correlated with spike RBD IgG (r, 0.9736) and nucleocapsid RBD IgG (r, 0.8740).

Stratified by baseline seropositivity status, those who were seropositive had longer median survival (29 vs 12 days; P =.003) and shorter median hospital stays (14.5 vs 34 days; P =.03). A total of 4 patients did not survive; all of who were seronegative at baseline.

Stratified by survivorship, median NT50 was 1535 among those who survived and 37.81 among those who died.

Longitudinal serum samples were assessed for 46 cytokines. A cluster analysis among all patients identified 3 distinct profiles of overall cytokine expression (high, moderate, and mild). For each individual patient over time, the cluster analysis identified 3 major disease trajectories (stable, increasing, and decreasing).

All patients who clustered into the stable disease trajectory and mild or moderate clusters survived, and all but 1 patient were seropositive at baseline. All patients who clustered into decreasing disease trajectory survived. Patients who had an increasing disease trajectory were seronegative at baseline.

This study was limited by its small sample size and low power. These findings should be confirmed among a larger study population before practical clinical application.

These findings suggested baseline NT50 at hospital admittance for COVID-19 associated with hospital stay and mortality and longitudinal cytokine expression profiles were indicative of disease trajectory.


Thiede JM, Gress AR, Libby SD, et al. Immune profiling reveals early disease trajectories associated with COVID-19 mortality: a sub-study from the ACTT-1 trial. J Infect Dis. 2021;jiab035. doi:10.1093/infdis/jiab035