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

The RNA detection data from pseudovirus cultures showed that the LOD of Sansure 267 was slightly lower than that of Daan NAAT. The RNA detection data from clinical 268 samples showed that the LOD of Bioperfectus NAAT was slightly higher than that of 269 the three other reagents but still reached the detection sensitivity of 20 copies per PCR In the present study, we further compared four NAATs for the detection of The copyright holder for this preprint (which was not peer-reviewed) is the . https: //doi.org/10.1101 //doi.org/10. /2020 and Hybribio reagents were all close to 40, indicating that the concentration of the 286 three samples may be lower than the LOD of Bioperfectus. Therefore, the LOD Obtaining high-quality viral RNA from original clinical samples is crucial to ensure 296 the accuracy of the detection results by NAATs. Therefore, the poor quality of 297 extraction reagents and the unoptimized extraction process of viral RNA would affect 298 the detection results. In this study, the performance of NAATs was evaluated by using 299 the same viral RNA samples. Therefore, the difference in the experimental results 300 only reflected the difference in the performance of nucleic acid amplification for 301 NAATs but not the diversity in the performance of the whole SARS-CoV-2 detection 302 system, which usually includes the nucleic acid extraction system. In summary, the four SARS-CoV-2 NAATs commonly used with commercial assays 313 showed good reliability, but the actual LOD was significantly higher than the declared 314 LOD. The analytic sensitivity of Daan, Sansure, and Hybribio NAATs was slightly The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.03.26.010975 doi: bioRxiv preprint