Sensitivity and performance of three novel quantitative assays for SARS-CoV-2 nuclear protein in blood

We compared the N measurement performance of three different commercial assays: Quanterix Simoa, Elecsys ECLIA, and Solsten ELISA. All three assays performed well in the viral culture dilution series (Spearman r > 0.99) but differed in LOD, with Quanterix Simoa being the lowest and Elecsys ECLIA the highest. In patient samples, the measured N concentrations differed in the three assays (Spearman 0.78–0.89). This nonlinearity is common when using patient samples in immunoassays due to interfering matrix effects from plasma. Since there are no established NIST calibrators yet, it is difficult to compare concentrations and cutoffs for each assay. However, the discriminatory power of each assay for finding positive and negative patients is more comparable. If two assays have different “N-positive” or “N-negative” categories, it is not possible to tell which result is more correct without a correct result, such as a PCR-confirmed diagnosis. Therefore, all 272 results of Quanterix Simoa and Solsten ELISA, Quanterix Simoa and Elecsys ECLIA, and Solsten ELISA and Elecsys ECLIA were compared pairwise. The resulting six 2 × 2 contingency tables are shown in the table. 3.

Comparison of Quanterix Simoa and Elecsys ECLIA (Table 3a), all discordant results (36) were negative with Elecsys ECLIA and positive with Quanterix Simoa, indicating reduced sensitivity of the Elecsys ECLIA assay. Fisher’s exact test shows differences between assays (P < 0.0001). Cohens Kappa shows substantial agreement (0.74), as does the McNemar test (P < 0.0001).

Comparison of Solsten ELISA and Elecsys ECLIA (Table 3b), most of the conflicting results (37 out of 38) were negative with Elecsys ECLIA and positive with Solsten, indicating reduced sensitivity of the Elecsys ECLIA assay. Fisher’s exact test shows differences between assays (P < 0.0001). Cohens Kappa shows substantial agreement (0.72), as does the McNemar test (P < 0.0001).

Comparison of Quanterix Simoa and Solsten ELISA (Table 3c), the discordant category has the same number of samples (18). This indicates that two assays detected the same number but exactly the same he is not a SARS-CoV-2 positive patient, or that one assay is superior in terms of both sensitivity and specificity. is shown. However, the comparison with the Elecsys ECLIA assay is similar for his two assays, so the previous explanation seems most plausible. This conclusion is supported by the accompanying statistical analysis. Fisher’s exact test shows a difference between assays (P < 0.0001), while both McNemar's test (P = 0.87) and his Cohens Kappa show substantial agreement (0.73).

Of the three assays, Quanterix Simoa stands out for its low detection limit and wide measurement range. This causes some discrepancies in the results for the lowest and highest concentrations of N.

The Elecsys ECLIA assay stands out with the highest LOD, resulting in samples with N concentrations below the LOD being detected as negative, resulting in reduced sensitivity.

However, the basic conclusion of the comparison is that each assay is different, as the current study has no validated truths to compare. Performance in terms of sensitivity/specificity, positive and negative predictive value should be compared against preferably the gold standard method (NAAT test) for each defined objective (screening, desegregation, neutralizing monoclonal antibody treatment) as performed. should be evaluated for each assay.Formerly at Solsten ELISA¹³ and Quanterix Simoa³.

The Elecsys ECLIA assay was tested with off-label use of plasma/serum and intended use of saliva or nasal mucus.However, we performed assay internal controls (Table 1, 2), and excellent correlations were found (table 2, Spearman > 0.99) with or without addition of extraction buffer between EDTA/heparin plasma. Therefore, we have shown that plasma samples can be measured by the Elecsys ECLIA assay.

The primary reason for testing the Elecsys ECLIA assay is its routine high-throughput applicability. This is in contrast to the Quanterix Simoa and Solsten ELISA assays. These assays currently have problems associated with communicating with laboratory information systems (LIMS) and therefore rely on manual procedures. report.

Additionally, the Elecsys ECLIA Assay is a random access test with a turnaround time of less than 60 minutes. In contrast, the Quanterix Simoa and Solsten ELISA assays are analyzed in batches with a turnaround time of 180 minutes. Therefore, depending on demand/number of tests, it is often only run once a day.

Although the Elecsys ECLIA assay had a cut-off value of approximately 0.40 COI (12 ng/L, similar to the Quterix Simoa result by testing viral cultures), clinical risk stratification of Covid-19 patients treated with Covid-19 We believe that it can be used effectively for mAbs as studies using the Quanterix Simoa assay found that mAbs with N > 1000 ng/L benefit the most⁶This correlates to approximately 7 COI (1000 ng/L) in the Elecsys ECLIA assay, well above the estimated cutoff of 0.40 COI (12 ng/L). In the Solsten ELISA assay, the 1000 ng/L result for Quanterix Simoa corresponds to approximately 855 ng/L, again well above the cutoff of 10.0 ng/L. Therefore, each of the tested assays should be suitable for use in selecting patients eligible for neutralizing antibody therapy.

The presence of N in plasma is, at least in part, a marker of active or recently active viral replication in patients, and N is associated with cell membranes and is nuclease-resistant, making it more susceptible to viral replication than NAAT. Could be a better marker.^FourTherefore, it can be detected even after active replication has stopped.^19,20Detection of N in blood can also be used for rapid screening of patients for COVID-19, e.g. in emergency departments. This is because N is only produced by patients suffering from a limited range of diseases.^1,12Antigenemia, as measured by N, can be used as a clinical tool to find the most important patients and those eligible for mAb therapy.^8,9,Ten,11N has also been observed to appear in the urine of COVID-19 patients as a strong marker of severe systemic disease.^{twenty one}Both the Quanterix Simoa and Solsten ELISA assays are now being retrospectively tested in larger patient cohorts.^3,Four,13.

The major limitations of this study are the lack of gold standard NAAT validation, the lack of international NIST calibrators, and the off-label use of Quterix Simoa and Elecsys ECLIA assays.

In summary, the comparison of the three N assays (Quanterix Simoa, Solsten ELISA, Elecsys ECLIA) with all assays showed different but excellent linearity in standardized inactivated SARS-CoV-2 virus preparations. I was.

For specific use in the treatment of indications with mAbs, the Quanterix Simoa N assay has been reported to have positive effects in patients with concentrations above 1000 ng/L.⁶The corresponding concentrations for the Solsten ELISA and Elecsys ECLIA assays are 855 ng/L and 7 COI (1000 ng/L) respectively. However, each assay should be evaluated individually for specific indications, including screening, monitoring, and eligibility for neutralizing antibody therapy. The dataset used in the study was (Supplementary information).