Abstract
Unpredictable fatal outcome of COVID-19 is attributed to dysregulated inflammation. Impaired early adaptive immune response leads to late-stage inflammatory outcome. The purpose of this study was to develop biomarkers for early detection of host immune impairment at first diagnosis from leftover RNA samples, which may in turn identify high risk patients. Leftover RNA samples of COVID-19 patients at first diagnosis were stored. Following prospective follow-up, the samples were shorted and categorized into outcome groups. Impaired adaptive T cell response (severity score) and Impaired IL-10 response (undetectable IL-10 in the presence of high expression of a representative interferon response gene) were determined by RT-PCR based assay. We demonstrate that a T cell response based 'severity score' comprising rational combination of Ct values of a target genes' signature can predict high risk noncomorbid potentially critical COVID-19 patients with a sensitivity of 91% (95% CI 58.7-99.8) and specificity of 92.6% (95% CI 75.7-99) (AUC:0.88). Although inclusion of comorbid patients reduced sensitivity to 77% (95% CI 54.6-92.2), the specificity was still 94% (95% CI 79.8-99.3) (AUC:0.82). The same for 'impaired IL-10 response' were little lower to predict high risk noncomorbid patients 64.2% (95% CI 35.1-87.2) and 82% (95% CI 65.5-93.2) respectively. Inclusion of comorbid patients drastically reduce sensitivity and specificity51.6% (95% CI 33.1-69.8) and 80.5% (95% CI 64.0-91.8) respectively. As best of our knowledge this is the first demonstration of a metric-based approach showing the 'severity score' as an indicator of early adoptive immune response, could be used as predictor of severe COVID-19 outcome at the time of first diagnosis using the same leftover swab RNA. The work flow could reduce expenditure and reporting time of the prognostic test for an earliest clinical decision ensuring possibility of early rational management.