Analytical performance evaluation of Lumipulse® SARS‐CoV‐2 antigen assay in 392 asymptomatic patients

Abstract Introduction Severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infection is one of the current public health care challenges. The main strategy adopted to prevent the spread of infection is the rapid identification of COVID‐19‐positive subjects. The aim of this study was to compare the performance of Lumipulse® antigen immunoassay with the real‐time RT‐PCR, the gold standard for the diagnosis of SARS‐CoV‐2 infection, in a strictly selected asymptomatic population. Materials and Methods A total of 392 consecutive oro‐nasopharyngeal swabs were collected from patients with no symptoms related to COVID‐19 at the Emergency Department of AORN Sant'Anna e San Sebastiano, Caserta, Italy to evaluate the analytical performance of Lumipulse® SARS‐CoV‐2 antigen compared to qualitative real‐time RT‐PCR in asymptomatic patients. Results Lumipulse® SARS‐CoV‐2 antigen assay shows an overall agreement rate of 97% with a sensitivity of 96% and a specificity of 98%, with a PPV and NPV of 97%. The sensitivity varies according to the cycle threshold (C t)‐value reaching 100% and 86% with 15 < C t < 25 and C t ≥ 25, respectively. The ROC analysis yielded an AUC value of 0.98, suggesting that the antigen test may accurately detect SARS‐CoV‐2. Conclusion Our data showed that Lumipulse® SARS‐CoV‐2 antigen assay might be an efficient tool in the identification and limitation of SARS‐CoV‐2 transmission in large asymptomatic populations.

of the pandemic, WHO has reported approximately 590,000,000 cases of SARS-CoV-2 infection, with over 6,500,000 deaths, and in Italy, about 21,500,000 cases and more than 170,000 deceases. 3,4 SARS-CoV-2 is a Coronavirus with a diameter of 80-120 nm, belonging to the order of Nidovirales, with a positive-sense single-stranded RNA genome sized from 26 to 32 kilobases which codify four main structural proteins: spike (S), nucleocapsid (N), membrane (M), and envelope (E). [5][6][7] The binding of the Spike protein to the angiotensin-converting enzyme 2 (ACE-2) receptor, widely present in epithelial cells, is the first step in SARS-CoV-2 infection. [8][9][10] After 2 years of the pandemic, it is now widely clear that the most useful tool for controlling the spread of SARS-Cov2 infection is the rapid identification of potentially infected individuals, may be the only strategic policy able to prevent the outbreak of new pandemic waves. Among the different analytical methods nowadays commercially available, the use of last-generation antigenic tests has become recently more and more attractive, especially considering their rapidity and low costs, even though the gold standard still remains the molecular diagnosis characterized by a high sensitivity and specificity (>95%). [11][12][13][14] As recently reported, the SARS-CoV-2 nucleocapsid protein (NP) is an ideal target for a viral antigen-based diagnosis, especially considering its high immunogenicity and its massive expression during the early phases of the infection. 7,[9][10][11][12][13][14][15] The aim of this study has been to evaluate the performance of Lumipulse® SARS-CoV-2 antigen test (Fujirebio, Inc.), an automated quantitative chemiluminescence enzyme immunoassay technology (CLEIA) used to detect SARS-CoV-2 NP in oro-nasopharyngeal swab samples of asymptomatic patients, and compare its performance with qualitative real-time RT-PCR as reference test, with the purpose to demonstrate how this strategy can be useful for the early detection of new cases in order to prevent new pandemic waves. October to December 2020. The median age was 59.6 years (range 1-101). Male patients were 53% (n = 208) and females were 47% (n = 184) and over 70% of them were older than 50 years (Table 1).

| Samples collection
Samples were obtained using cotton swabs and viral transport media in UTM1 (Copan Diagnostics). Specimens were collected at the time of admission and stored at +4°C until nucleic acid extraction. SARS-CoV-2 antigen test was performed within 2 h after swab collecting and within 24 h the results were confirmed by real-time RT-PCR assay.
All procedures followed were in accordance with the ethical standards and the study complies with the Declaration of Helsinki.
Since all the diagnostic activities described in this study were part of routine laboratory operations necessary for SARS-CoV-2 screening and diagnosis at the local facility, no patient informed consent and Ethical Committee were necessary. The authors were not involved in the sample collection and they had no access to patient identifying information.

| SARS-CoV-2 antigen detection assay
A quantity of 500 μL of viral transport media from each oronasopharyngeal swab was centrifuged at 3000 g for 10 min. The  that have been bound to nucleic acids are magnetized, transferred, and released via the specialized magnetic rods during the process of extraction. Viral RNA was eluted with 20 μL buffer and used for RT-PCR assay.

| SARS-CoV-2 RNA detection using real-time RT-PCR
The

| Statistical analysis
Descriptive statistics were used to describe general information of patients. Continuous data were presented as mean, standard deviation (SD), median and interquartile range (lower and upper quarter).
Categorical data were presented in numbers, percentages, and a 95% confidence interval (95% CI
The mean NP concentration among the PCR-positive samples was significantly higher than among PCR-negative samples   Table 5).
We finally divided the group with a higher C t -value into two subgroups, 25 < C t < 30 and C t ≥ 30, and we observed a concordance of 100% and 83.3% (95% CI, 72.1-94.5%), respectively. Furthermore, the median antigen concentration of C t ≥ 30 group was 0.25 pg/mL (range 0.06-118.67 pg/mL), while the median antigen concentration of 25 < C t < 30 group was 60.29 pg/mL (range 28.15-811.94 pg/mL; Table S3).

| DISCUSS ION
Identifying SARS-CoV-2-positive cases among the overall asymptomatic population is still now the more reliable tool to limit the The data obtained in our study show that Lumipulse® has high sensitivity and specificity (96% and 98%, respectively), even if its sensitivity seems to be C t -value dependent. Indeed, we noticed a decrease of sensitivity from 100% to 86% as the C t -value increases and this is clearly observable in the subgroup with C t ≥ 30, in which the sensitivity decreases up to 83%. Our data seem  Note: Sensitivity, specificity, AUC (area under the curve), negative predictive value (NPV), positive predictive value (PPV), positive likelihood ratio (+LR) and negative likelihood ratio (−LR), diagnostic odds ratio (DOR), and the corresponding confidence interval (95% CI) were calculated for Lumipulse® SARS-CoV-2 antigen assay using real-time RT-PCR as reference.
gene. These data might be explained by a phenomenon of "primers and/or probe mismatch", able to invalidate the RT-PCR results.
ROC curve analysis performed with the aim to determine the cut-off value of the Lumipulse test able to establish the higher cor- test in a large and not at-risk population. So, we cannot exclude that the performance of this test might largely differ in another population with different clinical characteristics. [19][20][21][22][23] In addition, we compared a quantitative assay as Lumipulse SARS-CoV-2 Ag with a qualitative one (real-time RT-PCR), but our purpose was only to evaluate the performance of Lumipulse test as a screening test in an asymptomatic population, avoiding any evaluation on its quantitative performance, as previously described. 12,15,[19][20][21][22][23][24] In conclusion, it is evident that the relevance of this study is considered as a potential model for preventing SARS-CoV-2 infection in large asymptomatic populations. Indeed, it is now clear that asymptomatic COVID-19 adult outpatients play a crucial role as transmitters of SARS-CoV-2 infection in the hospital environment and therefore their early detection may certainly be an effective strategy to prevent the recrudescence of new pandemic waves. Moreover, considering the more and more reduced resources destined for the public health facilities, the Lumipulse SARS-CoV-2 antigen test shows many advantages: easy to use (completely automatized), reduced analysis times (approximately 45 min), and lower operational costs. [17][18][19][20][21][22][23][24] In addition, its diagnostic performance not seems to be afflicted by the new SARS-CoV-2 variants, which recently appeared in the pandemic scenario, [25][26][27][28] although further studies will be necessary to confirm these data.

CO N FLI C T O F I NTE R E S T S TATE M E NT
No potential conflict of interest was reported by the author(s).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.