Comparison of the analytical and clinical sensitivity of thirty-four rapid antigen tests with the most prevalent SARS-CoV-2 variants of concern during the COVID-19 pandemic in the UK

Abstract

AbstractBackgroundThe continued emergence of SARS-CoV-2 variants of concern (VOC) requires timely analytical and clinical evaluation of antigen-based rapid diagnostic tests (Ag-RDTs) especially those that are recommended for at home use.MethodsThe limit of detection (LOD) of 34 Ag-RDTs was evaluated using the most encountered SARS-CoV-2 VOC viral isolates (Alpha, Delta, Gamma, Omicron BA.1, Omicron BA.5) and the wild type (WT). Clinical sensitivity was further evaluated for five Ag-RDT utilising retrospective samples (Alpha, Delta, Omicron BA.1) and one Ag-RDT utilising prospective clinical samples (Delta and Omicron BA.1).FindingsFor the WT, Alpha, Delta, Gamma and Omicron (BA.1) variants 22, 32, 29, 31 and 32 of the 34 Ag-RDTs evaluated met the World Health Organisations (WHO) target product profile (TPP), respectively. Of the 31 Ag-RDTs included for Omicron BA.5 evaluation 29 met the WHO TPP. Additionally, the LODs for samples spiked with Omicron BA.5 were significantly lower than all other VOCs included (p<0.001). In the retrospective clinical evaluation when comparing RNA copies/mL, the Ag-RDTs detected Alpha and Omicron (BA.1) more sensitively than the Delta VOC. Samples with high RT-qPCR Cts (Ct>25) resulted in reduced test sensitivities across all variants. We used linear regression to model the 50% and 95% LOD of clinical samples and observed statistically similar results for all tests. In the prospective clinical samples, the sensitivity was statistically similar for the Delta VOC 71.9% (CI 95% 53.3-86.6%) and Omicron VOC 84.4% (CI95% 75.3-91.2%).InterpretationTest performance differs between SARS-CoV-2 VOCs, and high sensitivity was achieved when testing the Omicron BA.5 VOC compared to the WHO Ag-RDT requirements. Continuous evaluations must be performed to monitor test performance.FundingThis work was funded as part of FIND’s work as a co-convener of the diagnostics pillar of the Access to COVID-19 Tools (ACT) Accelerator, including support from Unitaid (grant number 2019-32-FIND MDR), the government of the Netherlands (grant number MINBUZA-2020.961444), and the UK Department for International Development (grant number 300341-102). Funding was also obtained from the MRC for RLB and CGB. The Facilitating Accelerated Clinical Evaluation of Clinical Diagnostics for COVID-19 (FALCON C-19) study was funded by the UK National Institute of Health and Care Research (NIHR).