Performance of five rapid serological tests in mild-diseased subjects using finger prick blood for exposure assessment to SARS-CoV-2

Abstract

Objectives: Assess the performance of five SARS-CoV-2 rapid serological tests (RST) using finger prick (FP) blood on-site to evaluate their usability for exposure assessment in population-based seroprevalence studies. Study design: Since cross-reactivity with common cold human coronaviruses occurs, serological testing includes a risk of false-positive results. Therefore, the selected cohort for RST-validation was based on combined immunoassay (presence of specific antibodies) and RT-qPCR (presence of SARS-CoV-2) data. RST-performance for FP blood and serum was assessed by performing each RST in two groups, namely SARSCoV-2 positive (n=108) and negative healthcare workers (n=89). Differences in accuracy and positive and negative predictive values (PPV, NPV) were calculated for a range (1-50%) of SARS-CoV-2 prevalence estimates. Results: The OrientGene showed overall acceptable performance, with sensitivities of 94.4% and 100%, and specificities of 96.6% and 94.4%, using FP blood and serum, respectively. Although three RST reach optimal specificities (100%), the OrientGene clearly outperforms in sensitivity. At a SARS-CoV-2 prevalence rate of 40%, this RST outperforms the other tests in NPV (96.3%) and reaches comparable PPV (94.9%). Although the specificity of the Covid-Presto is excellent when using FP blood or serum (100% and 97.8%, respectively), its sensitivity decreases when using FP blood (76.9%) compared to serum (98.1%). Conclusions: Performances of the evaluated RST differ largely. Only one out of five RST (OrientGene) had acceptable sensitivity and specificity using FP blood. Therefore, the latter could be used for seroprevalence studies in a high-prevalence situation. The OrientGene, which measures anti-RBD antibodies, can be valuable after vaccination as well.