The Performance of Two Rapid Antigen Tests During Population-Level Screening for SARS-CoV-2 Infection

Background: SARS-CoV-2 antigen assays offer a rapid mean to diagnose and isolate infected individuals. However, their utility in population-level screening is unknown. Objectives: The performance of two antigen tests in detecting SARS-CoV-2 was assessed among individuals randomly selected in the community. Study Design: A prospective study that performed head-to-head comparison of two SARS-CoV-2 antigen assays. Individuals were recruited during community SARS-CoV-2 screening over 10 working days. Demographic and clinical data were collected. Standard Q COVID-19 Ag test, a point-of-care chromatographic assay, was conducted immediately, and then the sample was transported to the virology laboratory to perform PCR and the LIAISON SARS-CoV-2 Ag chemiluminesence immunoassay. Results: respiratory samples from 991 individuals were collected, and 62 were positive by PCR. Inconclusive PCR results were observed in 19 samples and were excluded. The median age of participants was 40.2 years (IQR 32.3–47.8), and 932 (94%) were males. Most (77.4%) of infections were asymptomatic. The sensitivity and the specificity of the LIAISON assay were 43.3% (95%CI 30.6–56.8) and 99.9% (95%CI 99.3–100). The Standard Q assay had lower sensitivity (30.6%, 95%CI 19.6–43.7) but similar specificity (98.8%, 95%CI, 97.8–99.4). Similarly, the LIAISON assay had higher positive predictive value (96.3%, 95%CI 81–99.9% vs. 63.3%, 95%CI, 43.9–80.1%). Both assays performed better in symptomatic patients and among samples with a low-cycle threshold (Ct < 25). Conclusion: In our setting of random community surveillance, rapid antigen testing of nasopharyngeal swabs by either LIAISON SARS-CoV-2 Ag (DiaSorin) or Standard Q COVID-19 Ag (SD Biosensor) was less sensitive to detecting SARS-CoV-2 than the TaqPath COVID-19 RT-PCR.