Abstract
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is responsible for the pandemic respiratory infectious disease COVID-19. However, clinical manifestations and outcomes differ significantly among COVID-19 patients, ranging from asymptomatic to extremely severe, and it remains unclear what drives these disparities. Here, we studied 159 sequentially enrolled hospitalized patients with COVID-19-associated pneumonia from Brescia, Italy using the VirScan phage-display method to characterize circulating antibodies binding to 96,179 viral peptides encoded by 1,276 strains of human viruses. SARS-CoV-2 infection was associated with a marked increase in immune antibody repertoires against many known pathogenic and non-pathogenic human viruses. This antiviral antibody response was linked to longitudinal trajectories of disease severity and was further confirmed in additional 125 COVID-19 patients from the same geographical region in Northern Italy. By applying a machine-learning-based strategy, a viral exposure signature predictive of COVID-19-related disease severity linked to patient survival was developed and validated. These results provide a basis for understanding the role of memory B-cell repertoire to viral epitopes in COVID-19-related symptoms and suggest that a unique anti-viral antibody repertoire signature may be useful to define COVID-19 clinical severity.