Agreement between two commercially available SARS-CoV-2 serologic tests among Italian healthcare workers

After national lockdown in the effort to defeat the COVID-19 outbreak, the gradual reopening of Italy was accompanied by a series of measures to limit the infection, including extensive epidemiological surveillance by means of antibody testing on selected risk groups.¹ The Regional Council of Lombardy Region (Northern Italy), one of the areas most affected by COVID-19 on a global scale, initially authorized DiaSorin as the only provider of SARS-CoV-2 serologic test for diagnostic and epidemiological purposes. The use of a single test allowed comparisons among results and the harmonization of epidemiological data, but it caused delays in the kit supply and lack of instrument access, limiting further implementation of epidemiological surveillance on a large scale.  

While serologic tests lack of sensitivity as diagnostic test (IgG antibodies appears after symptom onset, thus they are not sensitive in the period when the risk of transmission is higher), they are particularly valuable for purposes other than the diagnostic as they allow to assess the infection burden, to outline the transmission rate in certain settings, and to evaluate the effectiveness of control policies.²  

We evaluated two commercial kits designed to detect SARS-CoV-2 IgG: Diasorin LIAISON®SARS-CoV-2  IgG (chemiluminescent assay, CLIA, Diasorin, Italy)³ and Euroimmun Anti-SARS-CoV-2 ELISA IgG (enzyme-linked immunosorbent assay, Lubeck, Germany)⁴ in the sero-epidemiological screening of healthcare workers (HCWs) in a tertiary children’s hospital in Milan. Both tests received CE approval. Sera from venous blood withdrawal of 624 HCWs was tested by Diasorin and Euroimmun with a prevalence of SARS-CoV-2 IgG positive subjects of 4.49% (28/624; 95%CI 3.00-6.42) and 4.97% (31/624; 95%CI 3.40-6.98), respectively. The same sample has also been tested by means of a lateral flow immunochromatographic assay for qualitative detection of anti-SARS-CoV-2 IgM/IgG (2019-New Coronavirus IgM/IgG Rapid Test Casette, Beijing, Beier Bioengineering) with a prevalence of 1.12% (7/624; 95%CI 0.45-2.30).  

The agreement between the CLIA and ELISA kits was evaluated by computing the overall percentage of concordant results and Cohen’s κ coefficient with 95% confidence intervals.
Twenty-eight out of 624 were CLIA positive, while 31 were positive to ELISA. The overall agreement between the two was substantial, with a Cohen’s κ of 0.73 (95%CI 0.60-0.86). The two tests gave concordant results for 609 (97.6%) subjects, of which 22 were tested positive to both assays. 
Discordant results were found in 15 subjects: 6 scored negative to the Euroimmun test and positive to Diasorin, while 9 were Euroimmun positive and Diasorin negative.  

Among CLIA negative-ELISA positive subjects 3 reported previous symptoms, all of them were tested for SARS-CoV-2 by nasopharyngeal swab with negative results. Considering the CLIA positive-ELISA negative subjects, only one was symptomatic (no SARS-CoV-2 test was performed).  
Diasorin and Beier rapid antigenic tests showed concordant results in 603/624 subjects (96.6%), resulting in a fair overall agreement, with a Cohen’s κ of 0.39 (95%CI 0.18-0.59). Agreement between Diasorin LIAISON®SARS-CoV-2 IgG, Euroimmune Anti-SARS-CoV-2 ELISA (IgG), and Beier 2019-New Coronavirus IgM/IgG Rapid Test Casette assays is summarized in table 1. 

Data here presented showed a substantial agreement between Diasorin and Euroimmun in the routine practice, which suggests that both could be used in epidemiological investigation on large scale and to monitor the spread in high-risk settings. Facing with the choice of multiples serologic tests, national authorities have struggled to balance testing urgency against sensitivity and specificity concerns. A recently published model has brought to attention that effective surveillance largely depends on the frequency of testing and it is only marginally improved by high test sensitivity.⁵ In settings of extensive testing, control policies should therefore prioritize accessibility and testing frequency over sensitivity or data harmonization/consistency. While current studies focused on the performance characteristics of serologic assay, cost, feasibility and turnaround time are of great value in epidemiological investigations and need to be better evaluated with appropriate studies. 

Conflicts of interest: none declared.

References

1. Coronavirus, le misure adottate dal Governo. Available from: http://www.governo.it/it/coronavirus-misure-del-governo (last accessed: 11.07.2020)
2. EUnetHTA Joint Action 3 WP4. Rapid collaborative review on the current role of antibody tests for novel Coronavirus SARS-CoV-2 in the management of the pandemic. Available from: https://eunethta.eu/wp-content/uploads/2020/06/RCR_OT_01-_Antibody-tests-for-SARS-CoV-2_23-06-2020.pdf
3. DiaSorin. LIAISON® SARS-CoV-2 S1/S2 IgG. The fully automated serology test for the detection of SARS-CoV-2 IgG Antibodies. Available from: https://www.diasorin.com/sites/default/files/allegati/liaison_sars-cov-2_s1_s2_igg_m0870004366_b.pdf (last accessed: 11.07.2020)
4. Euroimmun. Anti-SARS-CoV-2 ELISA (IgG) Instruction for use. Available from: https://www.coronavirus-diagnostics.com/documents/Indications/Infections/Coronavirus/EI_2606_D_UK_A.pdf (last accessed: 11.07.2020)
5. Larremore DB, Wilder B, Lester E, et al. Test sensitivity is secondary to frequency and turnaround time for COVID-19 surveillance. medRxiv 2020. Available from: http://medrxiv.org/content/early/2020/06/27/2020.06.22.20136309.abstract