Introduction: Quantifying antibodies against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and neutralising antibodies may help to understand protection at the individual and population levels. Determination of neutralising antibodies using classical virus neutralisation tests (VNT) is considered the gold standard, but they are costly and time-intensive. Enzyme-linked immunosorbent assay (ELISA)-based surrogate VNTs (sVNT) or anti-SARS-CoV-2 spike protein receptor binding domain immunoglobulins (anti-S-RBD Ig) may be suitable alternatives to VNTs. We aimed to (a) explore the correlations between anti-S-RBD Ig, VNT, and sVNT measurements and (b) describe humoral immunity against SARS-CoV-2 after vaccination, natural infection, and vaccine breakthrough infection in healthy blood donors.
Methods: We measured total anti-SARS-CoV-2 Ig in 5714 serum samples from 2748 healthy individuals visiting the Swiss Red Cross Blood Donation Centre in Basel from 03/2020 to 04/2022. We used the Elecsys® Anti-SARS-CoV-2 immunoassay (Roche) against the N- and S-receptor binding domain (RBD) proteins. In a subset of 548 samples from 123 donors, we conducted sVNTs against the Wuhan wild-type SARS-CoV-2 (SARS-CoV-2 Neutralizing Antibodies Detection Kit; Adipogen™). In 100 samples from 40 donors, we correlated sVNT and VNTs against the wild-type (D614G WU1) virus. Surveys were sent to the blood donors to collect data on their SARS-CoV-2 infection and vaccination status. Using this data, donors were categorised as "vaccination only", "infection before vaccination", "post-vaccine breakthrough infection", and "natural infection only".
Results: Our longitudinal observation study cohort consisted of 50.7% males with a median age of 31 years (range 18-75 y). Anti-SARS-CoV-2 N protein positivity rates per month indicate 57.1% (88/154) of the cohort was infected up to 04/2022. No differences in seropositivity were found between sexes, age groups, blood types (AB0 or RhD), and cytomegalovirus serostatus. We observed a high correlation between anti-S-RBD Ig and inhibition percentage (Spearman's ρ = 0.92, Kendall's τ = 0.77, p <0.0001). We determined the sensitivity and specificity for the manufacturers' thresholds for detecting virus-neutralising effects and computed the "best" cut-off based on our real-world data. We categorised 722/1138 (63.5%) donors as vaccination only (82.3%), post-vaccine breakthrough infection (7.8%), infection before vaccination (5.8%), and natural infection only (4.2%). We observed a lower inhibition percentage in the natural infection-only group than in all other vaccinated groups. The infection before vaccination group had higher anti-S-RBD Ig titres after the first vaccine dose than the other vaccinated groups.
Conclusion: In total, 57.1% of healthy blood donors were infected with SARS-CoV-2, but natural infection without evidence of vaccination seems to result in substantially lower neutralising antibody levels. An estimate of antibody neutralisation may be helpful to assess reinfection risk. Total anti-S-RBD Ig correlates with surrogate virus neutralisation test results, a surrogate for neutralisation; therefore, we suggest that total anti-S-RBD Ig may estimate the level of neutralising antibodies. The threshold for protection from an unfavourable clinical outcome must be evaluated in prospective clinical cohorts.