Evaluated Ag-RDTs
Nineteen Ag-RDT based on lateral flow principle were evaluated in this study (Table 1): (1) ActivXpress+ COVID-19 Ag Complete Kit (Edinburgh Genetics Ltd), referred to as ActivXpress+, (2) Biocredit COVID-19 Ag (Rapidgen Inc.), referred to as Biocredit, (3) Bioeasy 2019-nCoV Ag (Shenzhen Bioeasy Biotechnology), referred to as Bioeasy, (4) Espline® SARS-CoV-2 (Fujirebio Diagnostics Inc.), referred to as Espline, (5) Genedia W COVID-19 Ag (Green Cross Medical Science), referred to as Genedia, (6) iChroma COVID-19 Ag Test (Boditech Medical Inc.), referred to as iChroma, (7) Innova SARS-CoV-2 Antigen Rapid (Innova Medical Group Ltd.), referred to as Innova, (8) Mologic COVID-19 Ag Test device (Mologic Ltd), referred to as Mologic, (9) NowCheck COVID-19 Ag test (Bionote Inc.), referred to as NowCheck, (10) Panbio™ COVID-19 Ag Rapid Test (Abbott Rapid Diagnostics), referred to as Panbio, (11) Rapid SARS-CoV-2 Antigen test card (Excalibur Healthcare Services), referred to as Excalibur, (12) Respi-Strip COVID-19 Ag (Coris Bioconcept), referred to as Respi-Strip, (13) SARS-CoV-2 Antigen Rapid Test Kit (Joysbio Biotechnology Ltd.), referred to as Joysbio, (14) SARS-CoV-2 Rapid Antigen Test (co-developed by SD Biosensor Inc and Roche Diagnostics, distributed by Roche Diagnostics), referred to as Roche, (15) Standard-F COVID-19 Ag FIA (SD Biosensor Inc), referred to as Standard-F, (16) Standard-Q COVID-19 (SD Biosensor Inc), referred to as Standard-Q, (17) Sure Status COVID-19 Antigen Card Test (Premier Medical Corporation), referred to as Sure-Status, (18) Coronavirus Ag Rapid Test (Zhejiang Orient Gene Biotech Ltd.), referred to as Orient, (19) Wondfo 2019-nCoV Antigen Test (Guangzhou Wondfo Biotech Co.), referred to as Wondfo. The selection of the Ag-RDT was based on expression of interest via the Infection Innovation Consortium (iiCON) and Foundation of New Diagnostics (FIND). Companies had no involvement in the design or reporting of the study.
SARS-CoV-2 serial dilutions and quantification of copy numbers
The SARS-CoV-2 isolate REMRQ0001/Human/2020/Liverpool was propagated in Vero E6 cells (C1008; African green monkey kidney cells), maintained in DMEM with 2% fetal bovine serum (FBS) and 0.05 mg/ml gentamycin. Ten-fold serial dilutions of SARS-CoV-2 stock were made starting from 1.0 x 106 pfu/ml to 1.0 x 102 pfu/ml using culture media as a diluent (DMEM with 2% FBS % and 0.05 mg/ml gentamycin). Two-fold dilutions were made below the ten-fold LOD dilution to refine the LOD. For quantification, viral RNA was extracted using QIAmp Viral RNA mini kit (Qiagen, Germany) according to the manufacturer's instructions. The genome copies/ml (gcn/ml) were calculated using the COVID-19 Genesig RT-qPCR kit (PrimerDesign, UK). RT-qPCR testing was carried out using the Rotor-Gene Q (Qiagen, Germany), with a ten-fold serial dilution of using quantified specific in vitro-transcribed RNA38. A total of five replicates were tested for each standard curve point and extracted RNA from each culture dilution was tested in triplicate, and the gcn/ml was calculated from the mean Ct value of these replicates.
Preparation of SARS-CoV-2 sample matrices and LOD testing protocol
Three types of sample matrices were tested 1) direct viral culture supernatant, 2) spiked dry swabs and 3) spiked wet swabs in Amies media.
For the direct viral culture matrix, a specific volume of the serial dilutions was added directly to the extraction buffers at a 1:10 ratio except for Respi-Strip which was added at 1:1 ratio with the extraction buffer following the IFU.
For dry swab testing, the proprietary nasopharyngeal (NP) or nasal (N) swabs included in each individual kit was used except for Respi-Strip, which does not include swabs, and the recommended Eswab (Copan, Italy) was used instead. To prepare the dry swab matrix, the swab was soaked in 1 ml of the virus culture dilution series for 6-8 seconds, followed by immersion in the prescribed amount of proprietary reaction buffer solution.
For the preparation of spiked wet swabs, Eswab in Amies media (Copan, Italy) was used across all tests. The swab was first immersed in the serial viral dilutions for 6-8 seconds, then placed into the Amies media to mimic the sample collection stage. Ag-RDTs were evaluated by then immersing the same swab into the extraction buffer, except for test Respi-Strip where 100µl of the Amies was mixed at 1:1 with the extraction buffer following its IFU.
For all Ag-RDTs and matrices, the sample volumes applied, and procedures were performed as specified in the test specific IFUs.
The LOD was defined as the lowest dilution at which all three replicates were positive. Every dilution was tested in triplicate and non-spiked culture media and Amies were used as negative controls. Results were interpreted by two operators, each blinded to the result of the other. If a discrepant result was obtained, a third operator read any discrepant tests for a 2/3 result.
Effect of swab absorbance and volume of extraction buffer in the LOD of dry swabs
We investigated the effect of the absorbance of the proprietary swabs and extraction buffer provided with the Ag-RDT kit in the LOD using dry swab, i.e. if a less absorbent swab and larger volumes of extraction buffer resulted in a poorer LOD on dry swab compared with direct culture supernatant. To compare the effectiveness of each NP and N swab to recover sample, the amount of liquid absorbed by the swabs was measured. Five replicates of each swab brand were immersed in culture media for 6-8 seconds, then taped on the inside of a 50ml centrifuge tube. These were then centrifuged for 5 minutes at 1000g and the amount of liquid released was measured using a micropipette.
The degree of correlation of the difference between LOD of dry swabs and direct culture for the same Ag-RDT with the volume recovered by swab type and volume of proprietary were investigated by Spearman’s correlation coefficient rho (ρ). Statistical significance was set at P <0.05.
LOD after 7 days at -80° C and one freeze-thaw cycle
After performing the LOD experiments, the viral culture dilutions were stored at –80 °C for 7 days and then the LOD experiments were performed again using the dry NP and N swabs. This would help to assess the use of stored clinical samples could be used to facilitate evaluation of Ag-RDTs.