Utility of Different Surrogate Enzyme-Linked Immunosorbent Assays (sELISAs) for Detection of SARS-CoV-2 Neutralizing Antibodies

Abstract

The plaque reduction neutralization test (PRNT) is a preferred method for the detection of functional, SARS-CoV-2 specific neutralizing antibodies from serum samples. Alternatively, surrogate enzyme-linked immunosorbent assays (ELISAs) using ACE2 as the target structure for the detection of neutralization-competent antibodies have been developed. They are capable of high throughput, have a short turnaround time, and can be performed under standard laboratory safety conditions. However, there are very limited data on their clinical performance and how they compare to the PRNT. We evaluated three surrogate immunoassays (GenScript SARS-CoV-2 Surrogate Virus Neutralization Test Kit (GenScript Biotech, Piscataway Township, NJ, USA), the TECO^® SARS-CoV-2 Neutralization Antibody Assay (TECOmedical AG, Sissach, Switzerland), and the Leinco COVID-19 ImmunoRank™ Neutralization MICRO-ELISA (Leinco Technologies, Fenton, MO, USA)) and one automated quantitative SARS-CoV-2 Spike protein-based IgG antibody assay (Abbott GmbH, Wiesbaden, Germany) by testing 78 clinical samples, including several follow-up samples of six BNT162b2 (BioNTech/Pfizer, Mainz, Germany/New York, NY, USA) vaccinated individuals. Using the PRNT as a reference method, the overall sensitivity of the examined assays ranged from 93.8 to 100% and specificity ranged from 73.9 to 91.3%. Weighted kappa demonstrated a substantial to almost perfect agreement. The findings of our study allow these assays to be considered when a PRNT is not available. However, the latter still should be the preferred choice. For optimal clinical performance, the cut-off value of the TECO assay should be individually adapted.

Keywords: PRNT; SARS-CoV-2; neutralizing antibodies; surrogate ELISA.

Figure 1

SARS-CoV-2 (RBD) IgG and PRNT antibody titers in follow-up serum samples of six individuals vaccinated with BNT162b2. neg. = negative; - not tested.

Figure 2

Distribution of test results of the examined SARS-CoV-2 antibody assays in relation to the PRNT results. The width of each violin reflects the number of generated results: wide = high number of results, narrow = low number of results; * “borderline” result.

Figure 3

Distribution of SARS-CoV-2 IgG II Quant test results (BAU/mL) in relation to the results of the PRNT conducted in parallel, including mean and standard deviation bars. PRNT test results ≥1:10 are shown as filled data point symbols and negative results are shown as empty data point symbols. Fifteen samples were negative tested in the SARS-CoV-2 IgG II Quant as they did not generate a signal. As a result of the logarithmic scale of the y-axis, they are not depicted in the figure. * “borderline” result.

Figure 4

Distribution of the inhibition (%) of each examined sELISA in relation to the PRNT conducted in parallel, including mean and standard deviation bars. PRNT test results ≥1:10 (“borderline” or positive results) are shown as filled data point symbols and negative results are shown as empty data point symbols. Error bars outside the axis limits are not shown. (A) = GenScript; (B) = TECO; (C) = Leinco assay; * “borderline” result.