Comparison of three molecular diagnostic assays for SARS-CoV-2 detection: Evaluation of analytical sensitivity and clinical performance

Background Currently, SARS-CoV-2 RNA detection using real-time reverse-transcription PCR (rRT-PCR) is the standard diagnostic test for COVID-19 infection. Various rRT-PCR assays are currently used worldwide, targeting different genes of the SARS-CoV-2. Here, we compared the analytical sensitivity and clinical performance (sensitivity and specificity) of Allplex SARS-CoV-2/FluA/FluB/RSV assay (Seegene), Standard M nCoV real-time detection kit (SD Biosensor), and U-TOP COVID-19 detection kit (Seasun Biomaterials) for SARS-CoV-2 detection. Methods Two hundred and forty-nine nasopharyngeal swab samples were evaluated to compare the clinical performance of the rRT-PCR assays. For the analytical performance evaluation, two RNA controls with known viral loads-SARS-CoV-2 RNA control and SARS-COV-2 B.1.351 RNA control-were used to investigate the potential impact of SARS-CoV-2 variants, particularly the B.1.351 lineage. Results Limits of detection ranged from 650 to 1300 copies/ml for rRT-PCR assays, and the mean differences in cycle threshold (C-t) values of the two RNA controls were within 1.0 for each target in the rRT-PCR assays (0.05-0.73), without any prominent C-t value shift or dropouts in the SARS-COV-2 B.1.351 RNA control. Using the consensus criterion as the reference standard, 89 samples were positive, whereas 160 were negative. The overall clinical performance of rRT-PCR assays was comparable (sensitivity 98.88%-100%; specificity 99.38%-100%), whereas the sensitivities of each target gene were more variable. Conclusions The three rRT-PCR assays showed comparable analytical sensitivity and clinical performance. The analytical and clinical sensitivities of each target gene were influenced more by the primer and probe design than the target gene itself.