ABSTRACT
BackgroundHigh number of SARS-CoV-2 antibody tests are available in different formats, detect different types of antibodies, and use different target proteins. Sensitivity of these tests varies and could be also related to clinical symptoms and testing time. MethodsSerum samples from 97 COVID-19 patients and 100 controls were tested with 9 antibody tests (SNIBE, Epitope, Euroimmun, Roche, Abbott, DiaSorin, Biosensor, LIPS N, and LIPS S-RBD). The results were analyzed in context of clinical data. FindingsPositivity rate was of tests was following: N-LIPS test (91.8% cases), Epitope (85.6%), Abbott and in-house LIPS S-RBD (both 84.5%), Roche (83.5%), Euroimmun (82.5%), DiaSorin (81.4%), SNIBE (70.1%), and Biosensor (64.9%). Agreement between tests varied (71-95%). Correlation between patient symptoms score and antibody value was test-dependent: varied from strongest in LIPS N ({rho}=0.41; p<0.001) to nonsignificant (LIPS S-RBD). Testing time from symptoms influenced sensitivity in some tests more than anothers. InterpretationSensitivity of tests varied highly and combination of different tests may improve it. Relation of results to symptoms and testing time was test-dependent. Thus, some antibody tests seems to be more sensitive to detect antibodies early and in asymptomatic patients than others. FundingStudy was funded by SYNLAB Estonia and Estonian Research Council grant PRG377. Research in contextO_ST_ABSEvidence before this studyC_ST_ABSHigh variation in COVID-19 antibody tests sensitivity has been described. Relation between antibody response and clinical cause has been found in some studies but not in others. Its known that antibody response is time dependent, however seroconversion medians varied in different studies. Added value of this studyWe confirmed that SARS-CoV-2 antibody response depends on clinical symptoms and time of testing, but we also found that this relation is dependent on test setup and viral antigens used in the tests. This may explain contradictory results of previous studies. Implications of all the available evidenceOur study has practical implications showing that not all antibody tests work uniformly well in symptomatic and asymptomatic cases and in different time periods from disease onset. This should be taken into consideration in clinical practice for diagnosing COVID-19 and in epidemiological studies evaluating the seroprevalence especially in asymptomatic population.
