Comparison of the analytical sensitivity of seven commonly used commercial SARS-CoV-2 automated molecular assays

Abstract

The SARS-CoV-2 pandemic has challenged molecular microbiology laboratories to quickly implement and validate diagnostic assays and to expand testing capacity in a short timeframe. Multiple molecular diagnostic methods received FDA emergency use authorization (EUA) and were promptly validated for use nationwide. Several studies reported the analytical and/ or clinical evaluation of these molecular assays, however differences in the viral materials used for these evaluations complicated direct comparison of their analytical performance. In this study, we compared the analytical sensitivity (lower limit of detection, LOD) of seven commonly used qualitative SARS-CoV-2 molecular assays: the Abbott Molecular RealTime SARS-CoV-2 assay, the NeuMoDx™ SARS-CoV-2 assay, the Roche Cobas®SARS-CoV-2 assay, the BD SARS-CoV-2 reagents for BD MAX™ system, the Hologic Aptima® SARS-CoV-2 assay, the Xpert Xpress SARS-CoV-2 test, and the GenMark ePlex SARS-CoV-2 test. The comparison was performed utilizing a single positive clinical specimen that was serially diluted in viral transport media and quantified by the EUA approved SARS-CoV-2 droplet digital PCR (ddPCR) assay. Replicate samples were prepared and evaluated for reproducibility across different molecular assays with multiple replicates per assay. Our data demonstrated that the seven assays could detect 100 % of replicates at a nucleocapsid gene concentration of (N1 = 1,267 and N2 = 1,392) copies/mL. At a one log less concentration, the Abbott, the Roche, and the Xpert Xpress assays detected 100 % of the tested replicates.

Keywords: COVID-19; LOD; SARS-CoV-2.