Clinical performance evaluation of SARS-CoV-2 rapid antigen testing in point of care usage in comparison to RT-qPCR

Abstract

Background: Antigen rapid diagnostic tests (RDT) for SARS-CoV-2 are fast, broadly available, and inexpensive. Despite this, reliable clinical performance data from large field studies is sparse.

Methods: In a prospective performance evaluation study, RDT from three manufacturers (NADAL®, Panbio™, MEDsan®, conducted on different samples) were compared to quantitative reverse transcription polymerase chain reaction (RT-qPCR) in 5 068 oropharyngeal swabs for detection of SARS-CoV-2 in a hospital setting. Viral load was derived from standardised RT-qPCR Cycle threshold (C_t) values. The data collection period ranged from November 12, 2020 to February 28, 2021.

Findings: The sensitivity of RDT compared to RT-qPCR was 42·57% (95% CI 33·38%-52·31%). The specificity was 99·68% (95% CI 99·48%-99·80%). Sensitivity declined with decreasing viral load from 100% in samples with a deduced viral load of ≥10⁸ SARS-CoV-2 RNA copies per ml to 8·82% in samples with a viral load lower than 10⁴ SARS-CoV-2 RNA copies per ml. No significant differences in sensitivity or specificity could be observed between samples with and without spike protein variant B.1.1.7. The NPV in the study cohort was 98·84%; the PPV in persons with typical COVID-19 symptoms was 97·37%, and 28·57% in persons without or with atypical symptoms.

Interpretation: RDT are a reliable method to diagnose SARS-CoV-2 infection in persons with high viral load. RDT are a valuable addition to RT-qPCR testing, as they reliably detect infectious persons with high viral loads before RT-qPCR results are available.

Funding: German Federal Ministry for Education and Science (BMBF), Free State of Bavaria.

Keywords: Antigen rapid diagnostic test; COVID-19; Clinical evaluation; PCR; Performance evaluation; SARS-CoV-2.

Fig. 1

Enrolment of antigen rapid diagnostic test results in the study. RDT: Antigen rapid diagnostic test. RT-qPCR: Quantitative reverse transcription polymerase chain reaction.

Fig. 2

Demographics of the study population compared to the general population of the hospital's catchment area. Study population (red and blue bars, n = 5 067) was compared to the general population of the hospital's catchment area Lower Franconia (black line, n = 1 317 619) as of December 31, 2019. Due to privacy reasons, one person with diverse gender was excluded from the figure. No data on population with diverse gender was available for Lower Franconia. Population data were obtained from Bavarian federal office for statistics

Fig. 3

Antigen rapid diagnostic test performance compared to quantitative reverse transcription polymerase chain reaction by manufacturer. Sensitivity, specificity, positive predictive value and negative predictive value of antigen rapid diagnostic tests from three manufacturers (nal von minden NADAL®, Abbott Panbio™, MEDsan®) in comparison to quantitative reverse transcription polymerase chain reaction, n = 5 056. RDT from the different manufacturers were conducted on different samples (806 NADAL®, 1 029 Panbio™, 3 221 MEDsan®). As performed on different samples, test performance can only be compared indirectly.

Fig. 4

Antigen rapid diagnostic test result in comparison to viral load. Viral load was determined by quantitative reverse transcription polymerase chain reaction (RT-qPCR, n=99), dotted line: viral load of 10⁶ SARS-CoV-2 RNA copies per ml. Due to limited sample volume two samples could not be retested using the reference RT-qPCR method. ****: p<0•0001 (Mann-Whitney U test) RDT: Antigen rapid diagnostic test C_t: Cycle threshold

Fig. 5

Sensitivity of antigen rapid diagnostic testing in relation to viral load. Viral load was determined by quantitative reverse transcription polymerase chain reaction (n=99)