Host transcriptomic profiling of COVID-19 patients with mild, moderate, and severe clinical outcomes

Characterizing key molecular and cellular pathways involved in COVID-19 is essential for disease prognosis and management. We perform shotgun transcriptome sequencing of human RNA obtained from nasopharyngeal swabs of patients with COVID-19, and identify a molecular signature associated with disease severity. Specifically, we identify globally dysregulated immune related pathways, such as cytokine-cytokine receptor signaling, complement and coagulation cascades, JAK-STAT, and TGF- β signaling pathways in all, though to a higher extent in patients with severe symptoms. The excessive release of cytokines and chemokines such as <i>CCL2</i>, <i>CCL22</i>, <i>CXCL9</i> and <i>CXCL12</i> and certain interferons and interleukins related genes like <i>IFIH1</i>, <i>IFI44</i>, <i>IFIT1</i> and <i>IL10</i> were significantly higher in patients with severe clinical presentation compared to mild and moderate presentations. Differential gene expression analysis identified a small set of regulatory genes that might act as strong predictors of patient outcome. Our data suggest that rapid transcriptome analysis of nasopharyngeal swabs can be a powerful approach to quantify host molecular response and may provide valuable insights into COVID-19 pathophysiology.