Assessment of serological tests for antibodies to different antigens of the SARS-CoV-2 virus: comparison of six immunoassays

Abstract

The new coronavirus SARS-CoV-2 has become a global challenge to medicine and, in particular, laboratory diagnostics. The study of the antibodies’ level to SARS-CoV-2 can be used as a confirmation test in the diagnosis of a disease, but it becomes of paramount importance in assessing population immunity resulting from a disease or vaccination, as well as in selection of convalescent plasma donors. The kits developed in our country and abroad for detecting antibodies to the SARS-CoV-2 virus differ both in the methods of testing and in the used coronavirus antigens to which the antibodies are directed. The aim of this study was to compare the diagnostic sensitivity and specificity of five kits for the detection of IgG antibodies to the SARS-CoV-2 virus, based on different diagnostic methods. Serum samples from 137 COVID-19 convalescents and 166 donors of blood and its components were examined. The control group consisted of 50 blood sera collected at the beginning of 2019 and 19 sera collected in 2018 (before the advent of the SARS-CoV-2 virus) and stored at -70 °C. Testing was carried out in analytical systems: rapid test “COVID-19 IgM/IgG Rapid Test (Colloidal Gold)” (China), on an automatic immunochemical analyzer Abbott Architect™ i2000 and kit “SARS-CoV2-IgG” (Abbot, Chicago , IL USA), by the chemiluminescence method using an automatic analyzer of the CL series and kits of the “Mindray” company (China) “SARS-CoV-2 IgM” and “SARS-CoV-2 IgG” and by the enzyme immunoassay method on the kits of the companies “Diagnostic Systems” Ltd (Russia, Nizhny Novgorod) “DS-IFA-ANTI-SARS-CoV-2-G”, “Xema” Ltd (Federal State Budgetary Institution “National Medical Research Center of Hematology” of the Ministry of Health of Russia) “SARS-CoV-2-IgG-IFA” and “Vector-Best” CJSC (Russia, Novosibirsk)” SARS-COV-2-IgM-IFA-BEST” and “SARS-COV-2-IgG-IFABEST”. When comparing the results of testing 137 plasma samples on the Vector-Best and Mindray kits for IgG antibodies, 127 samples were positive, 7 samples were negative on both kits, the discrepancy was 2.2%. In the study of IgM antibodies, 32.1% were positive, and 52.6% were negative in both kits. The discrepancy rate was 15.3%. Out of 166 samples, 1 serum (0.6%) was negative in 5 kits. On the Mindray kit, IgG antibodies to the antigens of the SARS-CoV-2 virus were detected in 165 samples (99.4%), on Vector-Best – in 164 sera (98.8%), on Diagnostic systems – in 151 (90.96%), on Xema – in 154 (92.8%), and on Abbott – in 155 samples (93.4%). At the same time, 135 (81.33%) samples were positive in all kits, while 30 samples had discordant results (18.07%), and in 9 sera, specific IgG was not detected in 2 or more kits. ROC analysis revealed a high diagnostic value of all tested kits (AUC from 0.908 to 0.998), which indicates a high quality of the separation model of positive and negative samples (p < 0.001). With the cut-off set by the manufacturers, the sensitivity and specificity ranged from 82.8% and 93.3% for the Diagnostic Systems kit to 99.4% and 95.8% for the VectorBest kit. The calculated correlation coefficients were higher between kits with a similar composition of the antigen used in the kits; therefore, it is better to monitor the dynamics of antibodies by diagnostic kits from the same manufacturer.

Nonlinear interdependence of the results of measuring anti-SARS-CoV-2 IgG levels using Abbott and Euroimmun test systems

“…Differences between test systems may be associated with post-translational changes in recombinant proteins used as antigens since these processes are not the same in prokaryotes and eukaryotes [54] . In addition at the stage of transferring the target analyzed protein to the solid phase during manufacturing the variability of the results obtained significantly increases as a result of the emerging conformational changes [55] .…”

“…Thus, ideally, for good consistency of laboratory data the monitoring of antibody levels dynamics should be carried out using the same test system or using just such test systems with a similar antigen composition given that the correlation coefficients of sensitivity and specificity are higher between test systems [ 40 , 54 ]. Unfortunately this is not always possible when a blood donor applies to different laboratories or when it is necessary to summarize a large amount of data from many sources to assess the epidemiological situation at the population level.…”