Performance comparison of two nucleic acid amplification systems for SARS‐CoV‐2 detection: A multi‐center study

Abstract Background Many rapid nucleic acid testing systems have emerged to halt the development and spread of COVID‐19. However, so far relatively few studies have compared the diagnostic performance between these testing systems and conventional detection systems. Here, we performed a retrospective analysis to evaluate the clinical detection performance between SARS‐CoV‐2 rapid and conventional nucleic acid detection system. Methods Clinical detection results of 63,352 oropharyngeal swabs by both systems were finally enrolled in this analysis. Sensitivity (SE), specificity (SP), and positive and negative predictive value (PPV, NPV) of both systems were calculated to evaluate their diagnostic accuracy. Concordance between these two systems were assessed by overall, positive, negative percent agreement (OPA, PPA, NPA) and κ value. Sensitivity of SARS‐CoV‐2 rapid nucleic acid detection system (Daan Gene) was further analyzed with respect to the viral load of clinical specimens. Results Sensitivity of Daan Gene was slightly lower than that of conventional detection system (0.86 vs. 0.979), but their specificity was equivalent. Daan Gene had ≥98.0% PPV and NPV for SARS‐CoV‐2. Moreover, Daan Gene demonstrated an excellent test agreement with conventional detection system (κ = 0.893, p = 0.000). Daan Gene was 99.31% sensitivity for specimens with high viral load (C t < 35) and 50% for low viral load (C t ≥ 35). Conclusions While showing an analytical sensitivity slightly below than that of conventional detection system, rapid nucleic acid detection system may be a diagnostic alternative to rapidly identify SARS‐CoV‐2‐infected individuals with high viral loads and a powerful complement to current detection methods.

int/). Rapid spread of SARS-CoV-2 and its variants had brought a huge burden to the global economy and medical system. 1,2 Early diagnosis, supportive care, isolation of infected patients, and their contacts are key to epidemic prevention and control. Rapid, sensitive, and inexpensive methods to detect SARS-CoV-2 infection are therefore urgently needed and ultimately to halt the development and spread of COVID-19. Conventional nucleic acid detection system is highly sensitive and specific (though not 100% for each), rapid and widely used for pathogen detection and is currently the main method for diagnosing COVID-19. 3 On the other hand, it has disadvantages, such as site restrictions for specialty laboratory zones, highly dependent on thermal circulator, multiple detection steps and long time consuming, which makes it difficult to meet the increasing needs of nucleic acid screening and rapid detection in outpatient and emergency departments. [4][5][6][7] In order to meet this challenge, many rapid nucleic acid detection systems for SARS-CoV-2 have been born one after another, the whole process of their detection time were significantly shortened, and the instrument miniaturization (part of the detection system are portable), with or without the characteristics of nucleic acid extraction, amplification and detection integration, as contrast to conventional SARS-CoV-2 nucleic acid detection systems. Since these methods used now have not met the requirements for point-of-care testing (POCT), they were temporarily named as "rapid nucleic acid testing" in Chinese expert consensus on the rapid nucleic acid testing of SARS-CoV-2. The analytical and clinical performance of these newly rapid nucleic acid testing for SARS-CoV-2 have been rarely evaluated based on the Chinese population thus far. Even if there are similar studies, they generally have some weaknesses, such as small scale of clinical validation trials and insufficient scientific data support. [8][9][10][11][12] Therefore, the objective of present study was to evaluate the clinical diagnostic performance of SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) compared with SARS-CoV-2 conventional nucleic acid detection system in a large cohort of oropharyngeal swabs samples.

| Study subjects
Oropharyngeal swabs specimens, which were obtained for rapid and conventional nucleic acid detection, collected from a total of 63,370 were collected from laboratory information management system records. The study design is displayed in Figure 1.

| Clinical specimens detected by SARS-CoV-2 rapid nucleic acid detection system (Daan Gene)
Nucleic acid extraction was performed after oropharyngeal swabs specimens were collected from truck drivers according to the Briefly, the supporting kit is a single tube and single copy packaging specification, and there is no need to prepare the reaction system.

| Clinical specimens detected by SARS-CoV-2 conventional nucleic acid detection system
The process of specimen collection and nucleic acid extraction was the same as described in Section 2.

| Suspected positive and positive specimens re-tested by Shenzhen Center for Disease Control and Prevention (CDC)
For the specimens suspected to be positive or positive in the initial test of port laboratory, conventional nucleic acid detection was carried out using both SARS-CoV-2 nucleic acid detection kits (Biogerm Medical Technology Co., Ltd. and Daan Gene Co., Ltd.) to re-test the specimens by Shenzhen CDC. According to the

Technical Guidelines for COVID-19 Laboratory Testing (Ninth Edition)
issued by China CDC, the case confirmed as positive in the laboratory was further sequenced to monitor the mutation of virus genome, provide experimental data for the change of COVID-19 nucleic acid detection reagent, vaccine research and development strategy, and also provide support for the work of epidemiological control and traceability.

| Diagnostic accuracy criteria
True positive cases in this study were confirmed by Shenzhen CDC True negative cases in this study were confirmed in reference to the Technical Guidelines for COVID-19 Laboratory Testing (Ninth Edition) issued by China CDC 13 : Two consecutive SARS-CoV-2 nucleic acid tests were negative.

| Comparison of diagnostic performance of SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) and SARS-CoV-2 conventional nucleic acid detection system
Clinical detection results of 63,352 oropharyngeal swabs by both systems were finally enrolled in this analysis. As described in Table 1, the sensitivity of SARS-CoV-2 conventional nucleic acid detection system was higher than that of SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) (0.979, 95% CI:

| Concordance between SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) and SARS-CoV-2 conventional nucleic acid detection system
In general, agreement between these two systems for individual

CoV-2 rapid nucleic acid detection system (Daan Gene) and SARS-CoV-2 conventional nucleic acid detection system
The distribution of C t values with respect to the discordant specimens between SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) and SARS-CoV-2 conventional nucleic acid detection system is displayed in Figure 2. Discordant specimens all had relatively high C t values (indicative of a lower viral concentration in the specimen), ranged from 33.67 to 44.26, with median of 38.14, IQR of 2.95. Note: SE = a/(a + c); SP = d/(b + d); PPV = a/(a + b); NPV = d/(c + d).

TA B L E 1 Diagnostic accuracy of SARS-CoV-2 rapid and conventional nucleic acid detection systems for SARS-CoV-2 detection
1 SARS-CoV-2 rapid nucleic acid detection system.

| Sensitivity of SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) in respect to clinical specimens with different viral load
The diagnostic ability of detection system is closely related to the virus load of the tested sample. We further divided the viral loads of samples by C t values and determined the sensitivity of rapid nu- UDG enzyme and dUTP enzyme; as well as a high-efficiency enzyme system that can improve the specificity and sensitivity of PCR reaction, including hot start Taq antibody enzyme and c-mmlv enzyme. 23 However, so far relatively few studies have compared the diagnostic performance between this testing system and the SARS-CoV-2 conventional nucleic acid detection system.
Overall, results of this study showed that the sensitivity of the SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) was slightly lower than that of the conventional detection system, but their specificity for SARS-CoV-2 was equivalent.
And the SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) had ≥98.0% PPV and NPV for the detection of SARS-CoV-2.
Moreover, the SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) demonstrated an excellent test agreement with the widely used conventional nucleic acid detection system (κ = 0.893, p = 0.000). The diagnostic sensitivity of a nucleic acid test kit is closely related to the viral load of the sample. 24 Discordant specimens in this study all had relatively high C t values (indicative of a lower viral concentration in the specimen), ranged from 33.67 to TA B L E 2 Concordance between SARS-CoV-2 rapid and conventional nucleic acid detection systems Note: OPA = (a + d)/(a + b + c + d); PPA = a/(a + c); NPA = d/(b + d).

F I G U R E 2
The distribution of Ct values with respect to the discordant specimens between SARS-CoV-2 rapid and conventional nucleic acid detection system with high viral load (C t < 35) and 50% for low viral load (C t ≥ 35). SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) performed less well at specimens with C t values ≥35; however, the reduction in sensitivity is relatively unimportant since high C t values probably indicate a low transmission risk. 25 Viral load tends to be higher at the beginning of infection, that is, when the virus is most infectious. [26][27][28] Epidemiologically, high viral load carriers are more likely to be super-

spreaders. In a preliminary clinical study shown that patients with
Ct values equal or above 34 do not excrete infectious viral particles and thus can be discharged from hospital care or strict confinement for non-hospitalized patients. 29 It is generally believed that the virus culture in samples with C t > 30 is negative, and these infected individuals are non-infectious. 30,31 Furthermore, the sensitivity displayed by SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) for specimens with C t values ≥35 was still substantially higher than that most of the SARS-CoV-2 rapid antigen tests.  34 reported that the overall poor sensitivity of the COVID-19 Ag Respi-Strip did not allow using it alone as the frontline testing for COVID-19 diagnosis, with an overall sensitivity of 30.2%. Therefore, although the sensitivity of SARS-CoV-2 rapid nucleic acid detection system is not as high as that of SARS-CoV-2 conventional nucleic acid detection system, SARS-CoV-2 rapid nucleic acid detection system is a rapid and simple detection method that can identify COVID-19 patients with high transmission risk and may be helpful to check the virus clearance rate of patients.

| CON CLUS IONS
In summary, although the sensitivity of SARS-CoV-2 rapid nucleic acid detection system is not as high as that of SARS-CoV-2 conventional nucleic acid detection system, SARS-CoV-2 rapid nucleic acid detection system is a rapid and simple detection method that can identify COVID-19 patients with high transmission risk. For emergency and symptomatic patients, or for large-scale population screening at customs, railway stations, airports, etc., it is a better choice to use the SARS-CoV-2 rapid nucleic acid detection system for early triage and rapid management of the suspected population, and then use the conventional nucleic acid detection system for confirmation. It is believed that with the development of rapid nucleic acid detection system, more stable and high-quality SARS-CoV-2 rapid nucleic acid detection technologies will emerge and become an indispensable force in clinical diagnosis and treatment and epidemic prevention and control.

AUTH O R CO NTR I B UTI O N S
All authors have made substantial contributions to all of the follow-

F I G U R E 3
Sensitivity of SARS-CoV-2 rapid nucleic acid detection system (Daan Gene) in respect to clinical specimens with different viral load