Comparative performance of four nucleic acid amplification tests for SARS-CoV-2 virus

Yujuan Xiong; Zhen-Zhen Li; Qizhen Zhuang; Yan Chao; Fei Li; Yi-Yuan Ge; Yi Wang; Pei-Feng Ke; Xianzhang Huang

When 19 or all 20 replicates were positive, the concentration was temporarily defined 140 as LOD, followed by 20 replicates per day for two consecutive days, yielding a total 141 of 60 results. If more than 95% of the results were positive, the concentration was 142 finally determined as the LOD of the reagent. The experiments were spread over 3 143 days so that the standard deviations reflect the performance of the assay over a range 144 of typical laboratory conditions but without a change in reagent lots. Accuracy. Ten positive reference materials and two negative reference materials from 155 the performance verification reference material kit were used to evaluate accuracy. Analytical interferences. To investigate the analytical interferences, three 168 interference reference materials from the performance verification reference material 169 kit were used, including 6 g/dL hemoglobin, 30 g/dL albumin, and the mix of 100 170 µg/mL ribavirin and 100 µg/mL azithromycin. All tests were repeated three times. all four systems were less than 4%, and the total CVs were less than 5% (Table 2) . Analytical interferences. An assessment of interferences is shown in Table 3 . The 232 bias of Ct value for the N gene and ORF1ab gene on all four systems was less than 233 8.5% at the following interfering substance concentrations tested: 6 g/dL hemoglobin, 234 30 g/dL albumin, 100 µg/mL ribavirin, and 100 µg/mL azithromycin.