Background:
Circulating SARS-CoV-2 RNA may represent a more reliable indicator of infection than nasal RNA, but RT-qPCR lacks diagnostic sensitivity for blood samples.

Methods:
A CRISPR-augmented RT-PCR assay that sensitively detects SARS-CoV-2 RNA was employed to analyze viral RNA kinetics in longitudinal plasma samples from nonhuman primates (NHP) after virus exposure; to evaluate the utility of blood SARS-CoV-2 RNA detection for COVID-19 diagnosis in adults cases confirmed by nasal/nasopharyngeal swab RT-PCR results; and to identify suspected COVID-19 cases in pediatric and at-risk adult populations with negative nasal swab RT-qPCR results. All blood samples were analyzed by RT-qPCR to allow direct comparisons.

Results:
CRISPR-augmented RT-PCR consistently detected SARS-CoV-2 RNA in the plasma of experimentally infected NHPs from 1 to 28 days post-infection, and these increases preceded and correlated with rectal swab viral RNA increases. In a patient cohort (n=159), this blood-based assay demonstrated 91.2% diagnostic sensitivity and 99.2% diagnostic specificity versus a comparator RT-qPCR nasal/nasopharyngeal test, while RT-qPCR exhibited 44.1% diagnostic sensitivity and 100% specificity for the same blood samples. This CRISPR-augmented RT-PCR assay also accurately identified COVID-19 patients with one or more negative nasal swab RT-qPCR result.

Conclusion:
Results of this study indicate that sensitive detection of SARS-CoV-2 RNA in blood by CRISPR-augmented RT-PCR permits accurate COVID-19 diagnosis, and can detect COVID-19 cases with transient or negative nasal swab RT-qPCR results, suggesting that this approach could improve COVID-19 diagnosis and the evaluation of SARS-CoV-2 infection clearance, and predict the severity of infection.