key: cord-0687491-3o14hg36 authors: Caputo, Valerio; Bax, Cristina; Colantoni, Luca; Peconi, Cristina; Termine, Andrea; Fabrizio, Carlo; Calvino, Giulia; Luzzi, Laura; Panunzi, Giorgia Gaia; Fusco, Claudia; Strafella, Claudia; Cascella, Raffaella; Battistini, Luca; Caltagirone, Carlo; Salvia, Antonino; Sancesario, Giulia; Giardina, Emiliano title: Comparative analysis of antigen and molecular tests for the detection of Sars-CoV-2 and related variants: a study on 4266 samples date: 2021-04-18 journal: Int J Infect Dis DOI: 10.1016/j.ijid.2021.04.048 sha: b4539f2ab46070fa0b4bc73b7ab43deee7b19cce doc_id: 687491 cord_uid: 3o14hg36 Objectives The present study compared the performance of Lumipulse G Sars-CoV-2 Ag kit with TaqPath COVID-19 RT-PCR CE IVD kit. Methods The study was conducted on 4266 naso-oropharyngeal swabs. Samples were subjected to antigen RT-PCR tests for detection of Sars-CoV-2 and related variants. Statistical analyses were conducted in R software. Results We found 503 positives (including 138 H69-V70 deletion carriers) and 3763 negatives by RT-PCR, whereas 538 positives and 3728 negatives were obtained by antigen testing. We achieved empirical and binormal AU-ROCs of 0.920 and 0.990, accuracy of 0.960, sensitivity of 0.866, specificity of 0.973, positive and negative predictive values of 0.810 and 0.980. We obtained a positive correlation between viral loads and antigen levels (R2 = 0.81), founding a complete concordance for high viral loads (log10copies/mL>5.4). Antigen levels >222 pg/mL were found to be reliable in assigning positive samples (p < 0.01). Concerning variant carriers, antigen test detected them with the same accuracy of other positive samples. Conclusions Molecular and antigen tests should be evaluated regarding the prevalence of the area. In case of low prevalence, antigen testing can be employed as a first-line screening for the timely identification of affected individuals with high viral load, also if carriers of Sars-CoV-2 variants. concordance for high viral loads (log 10 copies/mL>5.4). Antigen levels >222 pg/mL were found to be reliable in assigning positive samples (p<0.01). Concerning variant carriers, antigen test detected them with the same accuracy of other positive samples. Conclusions: molecular and antigen tests should be evaluated regarding the prevalence of the area. In case of low prevalence, antigen testing can be employed as a first-line screening for the timely identification of affected individuals with high viral load, also if carriers of Sars-CoV-2 variants. Keywords: COVID-19; Naso-oropharyngeal swabs; Antigen test; Viral load; Sars-CoV-2 Variants; Screening. The pandemic caused by Severe Acute Respiratory Syndrome Coronavirus 2 (Sars-CoV-2) dramatically affects health and quality of life worldwide (Strafella et al., 2020b (Strafella et al., , 2020a . To date, the gold standard test for Coronavirus-19 disease (COVID-19) diagnosis is represented by Real Time PCR (RT-PCR), which detects viral genome with high-sensitivity in naso-oropharyngeal swab fluids (Pascarella et al., 2020; Wiersinga et al., 2020) . However, pre-and analytical issues, including long turnaround time, limit its use in specialized laboratories. Thus, novel laboratory and point-of-care tests for the detection of viral antigens have been developed in order to provide faster responses to a large number of individuals and to counteract the diffusion of infection. On this subject, a chemiluminescence-based assay has been recently developed for the quantitative measurement of SARS-CoV-2 N protein in swab and saliva (Lumipulse G SARS-CoV-2 Ag kit, Fujirebio). To date, comparative studies showed its high reliability (Gili et al., 2021; Hirotsu et al., 2021 Hirotsu et al., , 2020 , although the general application of antigen testing deserves to be further investigated and validated on larger cohorts, taking into account also the current spreading of novel, more infectious Sars-CoV-2 variants. Hence, the study aimed to compare the performance of the quantitative antigen test Lumipulse G SARS-CoV-2 Ag kit (Fujirebio) and of the molecular test TaqPath COVID-19 RT PCR CE IVD kit (ThermoFisher Scientific). The study was conducted on 4266 samples obtained from 2426 individuals, which were enrolled in Scientific Institute for Research, Hospitalization and Healthcare (IRCCS) Santa Lucia Foundation from December 2020 to February 2021. We further analyzed the correlation among the viral loads obtained by RT-PCR and the antigen level reported as pg/mL, finding a positive correlation (R 2 =0.81, Figure1B). Given these data, we subdivided our samples according to viral load ranges (Table 1) and assessed the accuracy of antigen testing for each range. A complete concordance was observed for samples with high viral load (log 10 copies/mL >5,4), whereas samples with lower concentrations of virus showed decreased accuracy (Table1). Subsequently, we splitted the samples according to the antigen level in order to identify a reliable range in which positive samples are correctly assigned by antigen testing. As a result, antigen levels> 222 pg/mL showed the highest concordance rate (99%), indicating thereby the ability to correctly assign of positive samples (p< 0.01). At lower antigens levels, instead, a decreased concordance rate (1.87 to 222 pg/mL= 76% ; <1.87 pg/mL= 52%) was reported. Interestingly, samples carrying the H69-V70 deletion were detected by antigen test with the same accuracy of other positive non-variant samples (Table1). Overall, these results showed that samples with high viral load, including variant carriers, were successfully identified by antigen test, which reported high concentration results (Figure1B) . On the other hand, antigen detection revealed a reduced accuracy for samples with lower viral load, requiring the confirmation by RT-PCR. Considering these data and evidence from literature (Hirotsu et al., 2021 (Hirotsu et al., , 2020 , the use of molecular and antigen tests should be carefully evaluated regarding the prevalence of infection in a specific area. As shown by the high reliability in correctly identifying negative samples, quantitative antigen testing may be suitable as a first-line screening in low prevalence communities. In fact, antigen testing may allow a faster identification of individuals infected with a high viral load and/or carrying viral variants, which need to be timely identified and isolated in order to control viral spreading. Conversely, antigen test has a limited utility in monitoring the infection in positive patients and in high prevalence communities. As several positive individuals with variable viral loads are expected in these areas, molecular testing for Sars-CoV-2 infection should be recommended to avoid false negative results. Correlation between antigen levels (pg/mL) and viral loads obtained by RT-PCR (log 10 copies/mL) A B J o u r n a l P r e -p r o o f Two-step strategy for the identification of SARS-CoV-2 variant of concern 202012/01 and other variants with spike deletion H69-V70, France Evaluation of automated test -2 antigen assay for detection of SARS-CoV-2 nucleocapsid protein (NP) in nasopharyngeal swabs for community and population screening Prospective Study of 1,308 Nasopharyngeal Swabs from 1,033 Patients using the LUMIPULSE SARS-CoV-2 Antigen Test: Comparison with RT-qPCR Comparison of automated SARS-CoV-2 antigen test for COVID-19 infection with quantitative RT-PCR using 313 nasopharyngeal swabs, including from seven serially followed patients COVID-19 diagnosis and management: a comprehensive review Investigation of Genetic Variations of IL6 and IL6R as Potential Prognostic and Pharmacogenetics Biomarkers: Implications for COVID-19 and Neuroinflammatory Disorders Genetic Variability among Populations Highlights a Possible Link with COVID-19-Related Neurological Complications Pathophysiology, Transmission, Diagnosis, and Treatment of Coronavirus Disease 2019 (COVID-19): A Review All authors read and approved the final version of the manuscript. The authors declare no funding The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.