Introduction

The COVID-19 pandemic, caused by severe acute respiratory coronavirus 2 (SARS-CoV-2) led to unprecedented, accelerated vaccine development (1) and expansive roll-out programs (2,3). Much of the global population now has some level of adaptive immunity to SARS-CoV-2 induced by exposure to the virus (natural infection), vaccination, or a combination of both (hybrid immunity).
Natural infection induced by, and/or vaccination against, SARS-CoV-2 leads to the development of both binding and neutralizing antibodies (nAbs) (4,5), and the induction of T-cell responses during active immune reaction and clearance of infection (6). Key questions that subsequently arise relate to the duration and the level of protection an individual might expect based on their infection and vaccination history. Studies of those infected early in the pandemic documented that natural SARS-CoV-2 infection afforded some level of protection against reinfection in most individuals, and that subsequent reinfections were typically less severe than the primary episode (Table 1 ). However, SARS-CoV-2 has high rates of mutation and heavily mutated variants have emerged (7). Most significant are the ‘variants of concern’ (VOCs) (8), and there is now ample evidence that protection against reinfection with the B.1.1.529/21K (Omicron) variant (9,10) is dramatically reduced compared with previous variants (Table 1 ).
Any descriptor of immunity based on patient history will encompass a population of individuals with vastly variable exposure to vaccines and viral variants with differing orders of immune challenge intensity. Unrecognised ‘silent infections’, especially in Omicron-positive subjects with underlying immunity, further complicate the assessment. Therefore derivation of potential immunity based on patient history requires assistance from a surrogate composite score to inform about protection and to aid decision making.

Correlates of protection or risk

In vaccinology, a correlate of protection (CoP) reflects a statistical non-causal relationship between an immune marker and protection after vaccination (11). Most accepted CoPs are based on antibody measurements (12) and vary depending on the clinical endpoint, for example protection from (symptomatic) infection or severe disease. In contrast, a correlate of risk (CoR) can be used as a measurement of an immunologic parameter that is correlated with a study endpoint (13) and can predict a clinical endpoint in a specified population with a defined future timeframe. Notably, antibody markers have been used as correlates of immune function in clinical trials of SARS-CoV-2 vaccine efficacy (14-19), and for identifying the risk of symptomatic infection by VOCs (20,21).
A CoR would likely comprise a measure of the immune component plus determinants that act to modify such a measure (a multi-component composite CoR). In general, the immune component of a composite CoR should be easily measured by widely available technologies that are amenable to automation, are scalable, cost-efficient, and have a rapid turn-around time. Given the relative complexity, cost and pre-analytic requirements for cellular immune response testing, the preferred candidate for the immune component of a CoR would be detection of humoral immune response(s) (i.e. antibody). This perspective evaluates the various elements that need to be accommodated in the development of an antibody-based composite CoR for reinfection with SARS-CoV-2 or severe COVID-19.