Longitudinal variation in SARS-CoV-2 antibody levels and emergence of viral variants: implications for the ability of serological assays to predict immunity. (preprint)

Background Serological assays are being deployed to monitor antibody responses in SARS-CoV-2 convalescents and vaccine recipients. There is a need to determine whether such assays can predict immunity, as antibody levels wane and viral variants emerge. Methods We measured antibodies in a cohort of SARS-CoV-2 infected patients using several high-throughput serological tests and functional neutralization assays. The effects of time and spike protein sequence variation on the performance and predictive value of the various assays was assessed. Findings Neutralizing antibody titers decreased over the first few months post-infection but stabilized thereafter, at about 30% of the level observed shortly after infection. Serological assays commonly used to measure antibodies against SARS-CoV-2 displayed a range of sensitivities that declined to varying extents over time. Quantitative measurements generated by serological assays based on the spike protein were better at predicting neutralizing antibody titers than assays based on nucleocapsid, but performance was variable and manufacturer positivity thresholds were not able to predict the presence or absence of detectable neutralizing activity. Even though there was some deterioration in correlation between serological measurements and functional neutralization activity, some assays maintained an ability to predict neutralizing titers, even against variants of concern. Interpretation The ability of high throughput serological assays to predict neutralizing antibody titers is likely crucial for evaluation of immunity at the population scale. These data will facilitate the selection of the most suitable assays as surrogates of functional neutralizing activity and suggest that such measurements may have utility in clinical practice.

Comparison of SARS-CoV-2 serological assays for use in epidemiological surveillance in Scotland (preprint)

Background: Sero-surveillance of SARS-CoV-2 is crucial to monitoring levels of population exposure and informing public health responses, but may be influenced by variability in performance between available assays. Methods: Five commercial immunoassays and a neutralising activity assay were used to detect antibodies to SARS-CoV-2 in routine primary care and paediatric samples collected during the first wave of the pandemic in NHS Lothian, Scotland as part of ongoing surveillance efforts. For each assay, sensitivity and specificity was calculated relative to consensus results and neutralising activity. Quantitative correlation was performed between serological and neutralising titres. Results: Seroprevalence ranged from 3.4-7.3 % in primary care patients and 3-5.9 % in paediatric patients according to different immunoassays. Neutralising activity was detectable in 2.8 % and 1.3 % respectively. Relative assay performance changed depending on comparison to immunoassay consensus versus neutralising activity and qualititative versus quantitative agreement. Cross-reactivity with endemic seasonal coronaviruses was confirmed by neutralising assay in false positives for one immunoassay. Presence of false positives for another assay was found specifically in paediatric but not adult samples. Conclusions: Five serological assays show variable accuracy when applied to the general population, impacting seroprevalence estimates. Assay performance may also vary in detection of protective neutralising antibody levels. These aspects should be considered in assay selection and interpretation in epidemiological studies.