A composite CoR: Antibody dynamics, serology in practice and challenges, and expert recommendations

The antibody component of a composite CoR should be developed under defined conditions. To provide insight into these conditions, an understanding of antibody dynamics is required.

SARS-CoV-2 antibody dynamics

Natural infection with SARS-CoV-2 elicits a diversity of antibodies including those targeting S and nucleocapsid (N) antigens (59,92) and the development of anti-RBD IgG antibodies is associated with improved patient survival (93). A detailed systematic review of 66 studies investigated antibody responses (94). Collectively, the evidence supports the induction of IgM production in the acute phase of natural infection (peak prevalence: 20 days) followed by IgA (peak prevalence: 23 days), IgG (peak prevalence: 25 days), and nAbs (peak prevalence: 31 days) after symptom onset (94).
Serum IgG has the longest half-life compared with the relatively transient IgA or IgM (95). A longitudinal analysis of 4558 individuals, measuring total anti-N antibodies, revealed that, whilst total antibodies begin to decline after 90–100 days, they may persist for over 500 days after natural infection (96). Specifically measuring nAb via plaque reduction neutralization test (PRNT) shows that infection yields a robust nAb response in most individuals (67). Some studies report that anti-S antibodies show greater persistence than anti-N antibodies (97,98).
Dramatic inductions of anti-S or anti-RBD IgG antibodies is indicative of vaccination (59,99,100). Primary vaccination by some vaccines, (but not all (101)), or boosters generates high nAb titers (100,102,103) or neutralizing responses (99). Notably, nAbs wane over time (21) with a half-life of 108 days (81) – although the level of decay may be assay or variant dependent (102) – and multiple clinical factors affect the duration of neutralization responses after primary vaccination (66).

Anti-SARS-CoV-2 antibody testing

Commercial high-throughput immunoassays

Numerous immunoassays for the detection of antibodies against SARS-CoV-2 are available, differing in the immunoglobulin class detected, target viral antigen, format, and output (qualitative, [semi]-quantitative) (reviewed in detail (104,105)).
Head-to-head comparisons from the pre-Omicron era reveal variable levels of performance between the assays (106-110), caused by numerous technical factors including assay methodology, format and antibodies used, timing of testing, and the targeted viral antigen. Comparison studies show that sensitivity for detecting prior infection by different serologic assays changes over time (111). Commercial assays developed early during the pandemic are based on ancestral/wild-type antigens. Subsequently, there is potential for differential performance in the Omicron-era: in particular, S- and RBD-specific immunoassays have shown significantly reduced performance (112-114), and decreased comparability of quantitative results (115).
Most common commercial immunoassays detect both binding and nAbs without differentiating between them, however certain assays measuring IgG or total antibodies correlate well with neutralizing capacity (14,78,116-122), acting as surrogates of neutralization. Cell-based virus neutralization tests can be used to measure neutralizing capability, but these are typically not readily available in clinical laboratories due to inherent test performance challenges associated with their methodology, time and cost (123).
Expert recommendations
Mature immune responses are dominated by IgG. Serologic assays that measure IgG or total antibodies (if skewed towards IgG) that correlate with neutralizing activity and focus on anti-RBD should be used for the serologic component of a composite CoR; anti-N antibodies are unlikely to be neutralizing as the N protein is located within the viral envelope (59).
Assays should be adapted for accurate measurement of the modified antigen, if applicable. However, frequent adaptation of assays is unlikely if several variants are circulating in parallel and due to regulatory requirements for assays. Therefore, studies are needed to determine assay applicability in the present conditions, especially since RBD mutations frequently occur and recombinant versions of RBD or S are commonly used in immunoassays (105). Accordingly, the upper and lower thresholds of any CoR may need modification.
External ring trials show poor comparability of assays from different manufacturers (124,125) and there are significant challenges with the current binding antibody units (BAU) standardization, due to multiple factors, including different assay methods, antibody class(es) detected and target antigen used. Of note, BAU reference materials were derived from UK convalescent individuals infected in 2020 (126) (pre-Omicron), and there are vastly different BAU standardized values (Kroidl et al 2023, submitted). Antibody measurements should be harmonized across assays from different manufacturers, irrespective of the different epitopes utilized, to reduce variability. To support this, there is an urgent need for external quality assessment, production of robust traceable certified reference materials, standards for different variants, and improved documentation of the methods on laboratory reports. Age-specific normalization of reference intervals in defined groups, by means of z-log transformation and documentation in antibody passes, may further improve the comparability of assays. Stakeholders should agree on minimum performance-based criteria to develop the gold standard for CoR, allowing validation of secondary assays.
Finally, systemic cellular assays could provide a comprehensive profile of the immune response, especially in immunocompromised and susceptible individuals who are not able to mount a robust antibody response. Currently, they lack scientific evidence and their use in clinical practice still remains uncertain.

Sample matrices

Systemic anti-SARS-CoV-2 antibody testing can be performed on blood, plasma/serum, or dried blood spots (DBS) (105,127,128); Wieser et al 2023, submitted). An advantage of whole blood or DBS collection is the ease in obtaining the sample. Whilst many methodologies focus on systemic testing, infection with SARS-CoV-2 or vaccination against COVID-19 induces mucosal antibodies (129,130), thus secretions such as saliva offer another possibility. Antibody dynamics will differ depending on the material in question (131), and sample types are subject to specific idiosyncrasies, such as additional pre-processing, that need to be accounted for (132). Currently secretion-based testing is less suitable for a composite CoR as performance is variable (133).
Expert recommendations
A composite CoR will likely be sample matrix-specific. Our preference is for plasma/serum, as this sample matrix has the largest evidence base, shows the least variability, experiences less interference than whole blood, and is consistent with CoRs established for other infectious diseases. DBS would be also possible, but variability is high, and few laboratories have an established workflow.

Serologic testing formats

Formats include high-throughput automated enzyme immunoassay/ electrochemiluminescence immunoassay/enzyme-linked immunosorbent assay (certified and used in central laboratories and hospitals), point-of-care (POC) testing (used in emergencies and outpatients setting), and direct-to-consumer testing (at-home use with online services). POC testing is gaining in popularity, but methodological variation is higher (134) and any method that relies upon sampling from untrained individuals is less reliable for (semi)quantitative measurements (135).
Expert recommendations
We recommend automated assays that are approved by location-specific regulatory agencies and performed in certified and centralized laboratories. Home sampling/DBS would contribute to a reduction in clinician workload, particularly in high-density residential facilities, but methods are not yet sufficiently robust. At this time, there is no clear benefit in POC testing as urgent results are not critical.

Frequency of sampling and optimal time point

Considering antibody dynamics, several important questions arise: what is the optimal time point for measurement; would the timing differ depending on the vaccine schedule, and/or the presence of previous infection of a specified severity; should antibody levels be measured once or serially? Whilst single values can be plotted into modelled curves showing decrease rates over time, serial measurements could further refine the composite CoR. Only individuals with symptomatic disease or vaccination are known to stabilise the curve — infections that are sufficiently mild to lack detection will impact the composite CoR model.
Expert recommendations
As most individuals have experienced infection or vaccination, and titers are generally high and more stable than with single exposures, sampling should be performed annually or less. Serologic evaluation should be conducted more frequently in the elderly or immunocompromised than the general population (time interval to be defined), depending on any underlying disease and/or treatment.