Antigen-Based Testing but Not Real-Time Polymerase Chain Reaction Correlates With Severe Acute Respiratory Syndrome Coronavirus 2 Viral Culture

Abstract

Background: Individuals can test positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) by molecular assays following the resolution of their clinical disease. Recent studies indicate that SARS-CoV-2 antigen-based tests are likely to be positive early in the disease course, when there is an increased likelihood of high levels of infectious virus.

Methods: Upper respiratory specimens from 251 participants with coronavirus disease 2019 symptoms (≤7 days from symptom onset) were prospectively collected and tested with a lateral flow antigen test and a real-time polymerase chain reaction (rt-PCR) assay for detection of SARS-CoV-2. Specimens from a subset of the study specimens were utilized to determine the presence of infectious virus in the VeroE6TMPRSS2 cell culture model.

Results: The antigen test demonstrated a higher positive predictive value (90%) than rt-PCR (70%) when compared to culture-positive results. The positive percentage agreement for detection of infectious virus for the antigen test was similar to rt-PCR when compared to culture results.

Conclusions: The correlation between SARS-CoV-2 antigen and SARS-CoV-2 culture positivity represents a significant advancement in determining the risk for potential transmissibility beyond that which can be achieved by detection of SARS-CoV-2 genomic RNA. SARS-CoV-2 antigen testing can facilitate low-cost, scalable, and rapid time-to-result, while providing good risk determination of those who are likely harboring infectious virus, compared to rt-PCR.

Keywords: SARS-CoV-2; SARS-CoV-2 live culture; antigen testing; rt-PCR testing; viral load.

Figure 1.

A, The 38 real-time polymerase chain reaction (rt-PCR) assay–positive specimens are plotted by log₁₀ copies/mL (y-axis) and are stratified by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) live culture results (negative, n = 10; positive, n = 28). The median and interquartile range (IQR) values for the rt-PCR–positive/SARS-CoV-2 TMPRSS2 culture–negative tests were 4.21 and 1.37, respectively; the median and IQR values for the rt-PCR–positive/SARS-CoV-2 TMPRSS2 culture–positive tests were 7.39 and 1.66, respectively. The mean values for the SARS-CoV-2 TMPRSS2 culture–negative and SARS-CoV-2 TMPRSS2 assay–positive specimen groups were significantly different (4.01 vs 7.16, respectively; P < .001 based on 2-sample t test [2-tailed]). Antigen test–positive results are indicated as red data points (n = 29), and the antigen test–negative results (n = 9) are indicated by the green data points. B, Probit models linking viral load to the probability of positive result of rt-PCR (Lyra), antigen test (Veritor), SARS-CoV-2 TMPRSS2 culture, and SARS-CoV-2 VeroE6 culture (refer to Huang et al [17]). Viral load levels at which there is a 5% chance of positive result: 1.6, 2.6, 4.5, and 5.75 log₁₀ copies/mL for rt-PCR, antigen, SARS-CoV-2 TMPRSS2 culture, and SARS-CoV-2 VeroE6 culture, respectively.