Sensitivity evaluation of 2019 novel coronavirus (SARS-CoV-2) RT-PCR detection kits and strategy to reduce false negative

The early detection and differential diagnosis of respiratory infections increase the chances for successful control of COVID-19 disease. The nucleic acid RT-PCR test is regarded as the current standard for molecular diagnosis. However, the maximal specificity confirmation target ORF1ab gene is considered to be less sensitive than other targets in clinical application. In addition, recent evidence indicated that the initial missed diagnosis of asymptomatic patients with SARS-CoV-2 and discharged patients with “re-examination positive” might be due to low viral load, and the ability of rapid mutation of SARS-CoV-2 also increases the rate of false-negative results. Moreover, the mixed sample nucleic acid detection is helpful in seeking out the early community transmission of SARS-CoV-2 rapidly, but the detection kit needs ultra-high detection sensitivity. Herein, the lowest detection concentration of different nucleic acid detection kits was evaluated and compared to provide direct evidence for the selection of kits for mixed sample detection or make recommendations for the selection of validation kit, which is of great significance for the prevention and control of the current epidemic and the discharge criteria of low viral load patients.


INTRODUCTION
The coronavirus that caused the outbreak was identified in the case of viral 48 pneumonia in Wuhan in 2019 [1][2][3], and was named 2019-nCoV/SARS-CoV-2 by the 49 World Health Organization (WHO) [2,4,5]. SARS-CoV-2 belongs to the coronavirus 50 genus β and its genome is single-stranded, non-segmented positive-sense RNA [6], 51 which is the seventh known coronavirus that can infect humans [1,7]. Similar to other 52 pathogenic RNA viruses, the genetic material RNA is the first marker to be detected. 53 Nucleic acid detection or sequencing is currently used in conjunction with pulmonary 54 CT for clinical diagnosis of 9]. As the course of the disease progresses, 55 antibodies IgM and IgG will be produce by the human immune system. Although, 56 antibody tests play a major role in monitoring the response to future immunization 57 strategies and demonstrating previous exposure/immunity, the antibody positive rate 58 often lags behind the nucleic acid detection [10][11][12], and cross-reactions existed in 59 SARS-CoV antigen with autoantibodies [13]. 60 Theoretically, fluorescence quantitative RT-PCR detection is widely used as the 61 molecular diagnosis standard for 15]. Lately, the analysis showed 62 that the pattern of viral load change in COVID-19 patients was similar to that in 63 patients with influenza, but different from that in SARS and MERS (whose viral load 64 peaked about 10 days after the onset of symptoms) [16][17][18][19]. At present, a large 65 number of rapid gene detection technologies have been developed in succession, 66 which has great value for the screening of potential infectors and virus detection. 67 However, with too much emphasis on the "fast" characteristic, it is bound to cause a 68 All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.  [21,22], the low load virus will still lead to false negative or 73 spontaneous negative signals of thermostatic technology.

74
In COVID-19 patients, RT-PCR detection could be positive as early as one day 75 before the onset of symptoms, while most COVID-19 patients cannot be detected 76 before premorbid because of the low copy number of the virus [2, [7, 17, 23]. In 77 addition, some discharged patients appearing "re-examination positive" situation is 78 also because of the persistence of a small number of viruses. Unfortunately, the 79 positive rate of RT-PCR detection of SARS-CoV-2 is only 30%-50% at present [24,25] 80 due to improper sample collection, storage, and error detection [26]. Furthermore, 81 once the target gene mutated or deleted, the test results will be invalid [27,28].

82
RT-PCR nucleic acid detection not only has a high false negative rate [29], but also 83 has a low sensitivity [30]. Currently, the approved nucleic acid detection kits of the 84 SARS-CoV-2 genome are based on the most conserved and specific open reading 85 frame 1ab (ORF1ab), Envelope protein (E) and nucleocapsid protein (N) [6,31,32].

86
Athough ORF1ab is the highest specificity confirmation target gene, but is considered 87 to be less sensitive than other targets in clinical application [33], so does the pattern of 88 ORF1ab positive reports cause missed tests? Is it feasible to report based on positive 89 N or E genes? Clinically, it is recommended that samples with suspicious results or 90 All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 5, 2020. . https://doi.org/10. 1101 single channel positive results should be re-examined with another manufacturer's kit 91 or method. However, what is the basis for choosing the validation kit? This is a 92 problem that needs to be solved.  Prevention (China's CDC) (http://www.chinacdc.cn/jkzt/crb/zl/szkb_11803/jszl_11815/202003/t20200309_214241.html). And 108 the two swabs from nasopharyngeal and oropharyngeal were inserted into one sterile 109 tube containing 3 ml of Virus preservation solution. In addition, environmental 110 specimens were collected from surface in direct contact with the patient, such as inner 111 side of the mask, phone, doorknob, bedside, and etc. Each surface was wiped with one 112 All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 5, 2020. . https://doi.org/10.1101/2020.04.28.20083956 doi: medRxiv preprint synthetic fiber swab, and then inserted the swab into a sterile tube listed above.  (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 5, 2020.

Sensitivity evaluation of SARS-CoV-2 detection kits 148
To verify the sensitivity of the kits, we took nasopharyngeal and oropharyngeal (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 5, 2020. . https://doi.org/10. 1101 presented in in Figure.1A. Our solution is as follows ( Figure.   were suspected to be infected with the SARS-CoV-2 were enrolled for RT-PCR, and 181 two positive cases and two suspicious cases were found (Table 2). Then, the 182 All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 5, 2020.    (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 5, 2020. . https://doi.org/10.1101  (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 5, 2020. . https://doi.org/10.1101 regions. Moreover, study found that a deletion of 382 nucleotides in the ORF8 gene (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 5, 2020.