Development and validation of IMMUNO-COVTM: a high-throughput clinical assay for detecting antibodies that neutralise SARS-CoV-2
Authors: Vaandergast R. et al.
Link to paper: https://www.biorxiv.org/content/10.1101/2020.05.26.117549v1
Journal/ Pre-Print: BioRxiv
Tags; Clinical, Diagnostics
1. This assay can be used at the in biosafety CL2, is high-throughput and detects neutralising SARS-CoV-2 Spike antibodies in contrast to other tests like ELISA.
2. Validation of IMMUNO-COVTM using blinded analysis of 230 serum samples associated with clinical data and 80 other samples previously tested with commercially available ELISA.
3. There is a strong correlation (>0.8) between the IMMUNO-COV™ assay and SARS-CoV-2 neutralization test, suggesting that the results are clinically comparable.
This paper describes the generation and validation of a high-throughput clinical assay for quantifying SARS-CoV-2 Spike neutralising antibody levels in serum samples. This assay is done in CL2 containment and employs a vesicular stomatitis virus (VSV) expressing SARS-CoV-2 spike (S) protein, that upon infection, causes syncytia (cell-cell fusion) in Vero cell monolayers transduced with Dual Split protein (DSP) luciferase and GFP reporters. Syncytium formation is primarily read-out as a luciferase signal which is inhibited in the presence of SARS-CoV-2-neutralising antibodies. The assay exhibited 100% specificity and the optimal recommended sample dilution is 1:100. Luciferase readout is optimal between 24 and 30 hours and samples do not need to be heat-inactivated before carrying out the assay.
Impact for SARS-CoV2/COVID19 research efforts
Develop diagnostic tools for SARS-CoV2/COVID19. The authors claim that this high-throughput quantitative assay would be a powerful tool to screen for samples to be used in SARS-CoV2 plasma therapy trials. The test would also be useful to evaluate immune responses to vaccine candidates.
· In vitro study
Strengths and limitations of the paper
Novelty: Development of a high-throughput clinical test that reliably measures SARS-CoV-2 Spike neutralising antibodies in human serum without the typical full sample dilution series and CL3 containment.
Standing in the field: This is not the first paper that describes an assay focusing on the detection of SARS-CoV-2 neutralising antibodies using a surrogate assay. This is a link to one such paper: https://www.researchsquare.com/article/rs-24574/v1 There are already several automated immunoassays and ELISA-based assays on the market but they all measure total SARS-CoV-2 antibodies. For a brief comparison of some assays, refer to this recently published review https://www.sciencedirect.com/science/article/pii/S1386653220302225?via%3Dihub.
Appropriate statistical analysis was carried out. Graph Pad Prism was employed for these analyses. Parametric comparisons were used when appropriate and non-parametric approaches were employed for non-normal datasets.
Viral Model used: Clinically isolated SARS-CoV-2 and VSV-SARS-CoV-2-S-D19CT (Vesicular stomatitis virus in which its G glycoprotein is substitute with a 19-amino-acid C terminal deletion mutant of SARS-CoV-2 spike protein).
Translatability: This assay has a high potential to be employed for screening the patients’ antibody response to various vaccine candidates and to evaluate donor eligibility in convalescent plasma therapies. It is highly correlated (>0.8) to the full VNT, therefore, is a good surrogate.
The study is comprehensive and data presented are accurate. It would have been nice to know a bit more about plasma and serum samples. It is reported that positive individuals are those who tested positive for SARSCov-2 by PCR, but time after symptom onset for the samples is not known. This is an important piece of information when evaluating the sensitivity of an assay.
There are also slight inconsistencies across experiments in regard to how many hours post infection the luciferase output is measured. For example, in figure 4A, luciferase activity is measured 24h post infection, where as Figure 5A, it is 27h post infection and yet, in figure 5D, it was measured 23h post infection. The FACs data for the presence of ACE2 on the Vero cells suggests that there is nonspecific binding happening (the isotype control is highly similar to the anti-ACE2 ab). Also, it is important to recognise that there might be antibodies that neutralise SARS2 via ACE2 independent mechanisms (example via CD145) and therefore this assay may not be sensitive to those antibodies and might therefore give a false negative for some people.