Does a humoral correlate of protection exist for SARS-CoV-2? A systematic review

Abstract

Background

A correlate of protection (CoP) is an immunological marker associated with protection against infection. Despite an urgent need, a CoP for SARS-CoV-2 is currently undefined.

Objectives

Our objective was to review the evidence for a humoral correlate of protection for SARS-CoV-2, including variants of concern.

Methods

We searched OVID MEDLINE, EMBASE, Global Health, Biosis Previews and Scopus to January 4, 2022 and pre-prints (using NIH iSearch COVID-19 portfolio) to December 31, 2021, for studies describing SARS-CoV-2 re-infection or breakthrough infection with associated antibody measures. Two reviewers independently extracted study data and performed quality assessment.

Results

Twenty-five studies were included in our systematic review. Two studies examined the correlation of antibody levels to VE, and reported values from 48.5% to 94.2%. Similarly, several studies found an inverse relationship between antibody levels and infection incidence, risk, or viral load, suggesting that both humoral immunity and other immune components contribute to protection. However, individual level data suggest infection can still occur in the presence of high levels of antibodies. Two studies estimated a quantitative CoP: for Ancestral SARS-CoV-2, these included 154 (95% confidence interval (CI) 42, 559) anti-S binding antibody units/mL (BAU/mL), and 28.6% (95% CI 19.2, 29.2%) of the mean convalescent antibody level following infection. One study reported a CoP for the Alpha (B.1.1.7) variant of concern of 171 (95% CI 57, 519) BAU/mL. No studies have yet reported an Omicron-specific CoP.

Conclusions

Our review suggests that a SARS-CoV-2 CoP is likely relative, where higher antibody levels decrease the risk of infection, but do not eliminate it completely. More work is urgently needed in this area to establish a SARS-CoV-2 CoP and guide policy as the pandemic continues.

About the Authors:

Julie Perry

Roles Conceptualization, Data curation, Formal analysis, Project administration, Writing – review & editing

Affiliations Public Health Ontario, Toronto, Ontario, Canada, Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada

Selma Osman

Roles Data curation, Formal analysis, Writing – review & editing

Affiliation: Public Health Ontario, Toronto, Ontario, Canada

James Wright

Roles Data curation, Formal analysis, Writing – review & editing

Affiliation: Public Health Ontario, Toronto, Ontario, Canada

Melissa Richard-Greenblatt

Roles Conceptualization, Methodology, Writing – review & editing

Affiliation: Public Health Ontario, Toronto, Ontario, Canada

Sarah A. Buchan

Roles Conceptualization, Methodology, Writing – review & editing

Affiliations Public Health Ontario, Toronto, Ontario, Canada, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada, Centre for Vaccine Preventable Diseases, University of Toronto, Toronto, Ontario, Canada

Manish Sadarangani

Roles Conceptualization, Methodology, Writing – review & editing

Affiliations Department of Pediatrics, University of British Columbia, Vancouver, British Columbia, Canada, Vaccine Evaluation Center, BC Children’s Hospital Research Institute, Vancouver, British Columbia, Canada

Shelly Bolotin

Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Supervision, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliations Public Health Ontario, Toronto, Ontario, Canada, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada, Centre for Vaccine Preventable Diseases, University of Toronto, Toronto, Ontario, Canada, Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada

ORCID logo https://orcid.org/0000-0002-9100-5928

Introduction

Both previous infection and vaccination against SARS-CoV-2 provide protection against infection and severe disease, but the mechanism and durability of that protection remains unclear [1]. Immunity to SARS-CoV-2 is likely both humoral and cellular [2], but it is uncertain whether a correlate of protection (CoP) for SARS-CoV-2 exists, and if so, whether it is easily quantifiable using diagnostic testing. Without a CoP, serological testing cannot confirm immunity, leaving an evidence gap in public health policy particularly as new variants of concern emerge.

A CoP is an immunological marker associated with protection from an infectious agent following infection or vaccination [3]. Some CoPs are mechanistic (i.e. directly responsible for protection), while others are non-mechanistic or surrogate, and although not directly responsible for protection, can be used in substitute of the true correlate [3, 4]. A CoP can be absolute, where protection against disease is certain above a threshold, or relative, where higher levels of a biomarker correspond to more protection [2]. Some correlates vary by endpoint (e.g. symptomatic infection or severe disease), or are only applicable to a specific endpoint [3]. The majority of CoPs described are humoral and used in a surrogate manner, as these antibodies are easier to detect in clinical laboratory settings than components of cellular immunity [5].

Elucidating a CoP for SARS-CoV-2 is critical for improving our understanding of the extent and duration of protection against infection for individuals and populations. At the individual level, a CoP would provide clear immunological vaccine trial endpoints, and therefore may provide a pathway to licensure for new vaccines [5]. If measurable using a diagnostic test, a CoP would enable determination of individual and community-level immunity, which is particularly important for immunocompromised individuals [6, 7] and the assessment of population level immunity through serosurveys [5].

The search for a SARS-CoV-2 CoP is further complicated by the emergence of variants of concern (VOCs). Sera from previously infected and/or vaccinated individuals have reduced neutralizing ability against VOCs including Beta (B.1.351), Delta (B.1.617.2) and Omicron (B.1.1.529) [8–10], with the latter showing the greatest extent of immune evasion of all VOCs thus far [11]. This variation raises the possibility that a SARS-CoV-2 CoP may be VOC-specific.

With these facts in mind, and considering that an easily measurable CoP would most likely be humoral and not cellular, we performed a systematic review to assess the evidence for a humoral CoP for SARS-CoV-2.

Methods

Data sources and searches

We searched the OVID MEDLINE database from inception to December 31, 2021, and the EMBASE, Global Health, Biosis Previews and Scopus databases from inception to January 4, 2022. We used the NIH iSearch COVID-19 Portfolio tool to search for preprint articles published up to December 31, 2021. Our search included studies reporting either re-infection or breakthrough infection following vaccination. Full search terms used are reported in S1 Table. We also searched reference lists for suitable articles, and requested article recommendations from experts in the field.

Study selection

One reviewer screened titles and abstracts using Distiller SR (Ottawa, Ontario, Canada). Studies passed title and abstract screening if their abstracts discussed re-infection with SARS-CoV-2 or breakthrough infection following vaccination; mentioned antibody measures specific to SARS-CoV-2; or mentioned a correlate or threshold of protection against SARS-CoV-2. We excluded studies that focused on immunocompromised populations or animal models.

Two reviewers screened full texts of articles that passed title/abstract screening using defined criteria (Table 1). We included studies reporting a quantitative CoP against SARS-CoV-2, and studies reporting re-infection or breakthrough infection along with associated pre-infection measures. If these studies reported aggregate antibody measures (i.e. geometric mean titres (GMT)) we required them to include summary statistics (i.e. statistical significance testing or 95% confidence intervals (95% CI)). We also included studies that correlated antibody levels to vaccine efficacy (VE) or effectiveness, but only if they provided statistical summary measures (e.g. a correlation co-efficient describing the relationship between antibody level and VE), or if they correlated an antibody concentration to a VE of 100% (i.e. absolute protection). We only included studies written English or French. We calculated a Cohen’s Kappa to assess inter-rater agreement for full-text screening.

[Figure omitted. See PDF.]

Table 1. Definitions applied to determine cases of re-infection and breakthrough in this systematic review.

https://doi.org/10.1371/journal.pone.0266852.t001

Data extraction and quality assessment

Two reviewers extracted data in duplicate from articles that met full-text screening criteria, using WebPlotDigitizer [12] for figures. We used the National Institutes of Health National Heart, Lung and Blood Institute (NIH NHLBI) Study Quality Assessment tools to assess study quality [13], adapting it by adding questions specific to this study. Studies correlating VE to antibody levels were evaluated using the Cohort and Cross Sectional Tool.

Data synthesis and analysis

We reported our results using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 [14]. A PRISMA reporting checklist can be found in the Supplemental files section (S1 Checklist). Recognizing that that the immune response following natural infection and vaccination may differ, we grouped studies involving re-infection separately from studies examining breakthrough infection.

Results

We identified 11,803 records for screening (Fig 1). After de-duplication, we screened 4,919 peer-reviewed studies, 783 preprint studies and 16 studies identified through expert recommendations and scanning of article reference lists. After title/abstract screening, full-text screening (Kappa = 1.0) and quality assessment, we included 25 articles in our review. Of these, 14 described SARS-CoV-2 re-infection along with individual or aggregate humoral measures [15–28], and 11 studies described SARS-CoV-2 breakthrough infection following vaccination or statistical modelling to explore associations between VE and antibody levels [29–39] (Table 2). Only two studies estimated a SARS-CoV-2 antibody CoP, both using statistical modelling methods [33, 34].

[Figure omitted. See PDF.]

Fig 1. PRISMA diagram.

https://doi.org/10.1371/journal.pone.0266852.g001

[Figure omitted. See PDF.]

Table 2. Summary of articles included in this review following re-infection and breakthrough infection definition screening, and types of evidence they describe.

https://doi.org/10.1371/journal.pone.0266852.t002

Studies describing SARS-CoV-2 re-infection

Fourteen studies met our SARS-CoV-2 re-infection definition and provided pre-infection antibody values (Table 3). These included seven cohort studies [15, 17, 18, 20–22, 27], and seven case reports [16, 19, 23–26, 28]. The majority of studies reported re-infection in healthcare workers, patients, or long term care home residents [15–18, 20, 22, 24–28], with a minority reporting re-infection in the general population [19, 21, 23]. When reported, specimen collection occurred between 14 days and seven months after initial infection [16, 26] and between 4 days and seven months before re-infection [20, 27]. Antibody testing methods included various commercial and laboratory developed enzyme-linked immunosorbent assays (ELISAs) targeting anti-spike (anti-S), anti-receptor binding domain (anti-RBD) and anti-nucleocapsid (anti-N) antibodies, as well as neutralization assays. No study utilized the World Health Organization (WHO) International Standard (IS) [40]. Only three papers reported on the SARS-CoV-2 lineage of the re-infection [16, 24, 26], with none reporting serological measures preceding re-infection with VOCs.

[Figure omitted. See PDF.]

Table 3. Articles describing SARS-CoV-2 re-infection along with individual or aggregate humoral measures#.

https://doi.org/10.1371/journal.pone.0266852.t003

Two studies compared antibody levels between re-infected and protected individuals. Krutikov et al. found no statistically significant difference in anti-N IgG between cases and controls (p = 0.544) but showed that individuals who were antibody-negative at baseline were at greater risk of infection than those who were antibody-positive [20]. Using Poisson regression, Lumley and colleagues also found that anti-S positive individuals were less likely to be infected compared to those who were anti-S negative (incidence rate ratio (IRR) of 0.11 (95% CI 0.03, 0.44)) [22]. Similar findings were observed using anti-N antibody (IRR = 0.11 (95% CI 0.03, 0.45)). Analysis of the association between continuous antibody concentrations and incidence was also statistically significant for both antibodies (p<0.001) [22].

Studies reporting antibody measures related to breakthrough infection or VE

We included 11 studies describing breakthrough SARS-CoV-2 infection. These included two case reports [36, 37], one cohort study [35], two case-control studies [29, 38], and two studies that re-analyzed antibody data from a clinical trial [32, 39]. Five in silico studies utilized statistical methods to explore the association between antibody levels and VE [30–34]. The populations studied were either clinical trials or other vaccine study participants [30–34, 39] or healthcare workers [29, 35–38]. Three studies reported results in WHO IS units (binding antibody units (BAU)/mL) [32, 33, 37].

Of the 11 studies describing breakthrough infection, six studies provided individual or aggregate humoral measures [29, 35–39], four studies used statistical modelling to explore associations between VE and antibody levels [30, 31, 33, 34], and one study included both humoral measures and statistical modelling [32] (Tables 4 and 5). Five studies [29, 36–39] reported the lineage of the breakthrough infection, and two modeling studies include VOCs in their analysis [30, 33].

[Figure omitted. See PDF.]

Table 4. Articles describing breakthrough following SARS-CoV-2 infection along with individual or aggregate humoral measures#.

https://doi.org/10.1371/journal.pone.0266852.t004

[Figure omitted. See PDF.]

Table 5. Articles describing statistical modelling to explore associations between VE and antibody levels#.

https://doi.org/10.1371/journal.pone.0266852.t005

Studies describing breakthrough infections following SARS-CoV-2 vaccination.

Seven of 11 studies reported antibody levels following one [35] or two doses of COVID-19 vaccine, including BNT162b2 (Pfizer-BioNTech) [29, 35–38] mRNA-1273 (Moderna) [32] and ChAdOx1 nCoV-19 (AstraZeneca) [39] (Table 4). Sera were collected between nine [36] and 109 days [32] after the second vaccine dose, but the time from sampling to breakthrough infection was not always reported. Antibody levels were assessed using a variety of commercial serology assays and/or neutralization assays. Five studies reported the viral lineage, including three studies reporting Alpha (B.1.1.7) [29, 37, 39], one reporting B.1.525 [36], and one reporting Delta (B.1.617.2) [38] infections.

Four studies compared aggregate antibody levels between cases and non-cases. Gilbert et al. calculated geometric mean concentration (GMC) ratios of cases to non-cases, ranging from 0.57 (95% CI 0.39, 0.84) to 0.71 (95% CI 0.54, 0.94), depending on antibody target and sampling interval [32]. Using Cox regression, the authors found statistically significant associations between increasing antibody levels and decreasing risk of COVID-19. Bergwerk et al. applied generalizing estimating equations to predict antibody levels and generate GMT ratios of cases to non-cases. For neutralizing antibodies, these ranged from a case-to-control ratio of 0.15 (95% CI, 0.04, 0.55) within the first month after the second vaccine dose to 0.36 (95% CI 0.17, 0.79) by the week before breakthrough infection [29]. Using linear regression, this study demonstrated a statistically significant correlation between cycle threshold (Ct) value of cases and neutralizing antibody level, suggesting an inverse relationship between antibody level and viral load. Feng and colleagues found no statistically significant difference between median antibody levels of cases and non-cases [39]. However, using a generalized additive model, symptomatic infection risk was found to be inversely correlated to antibody levels. Yamamoto et al. found no statistically significant difference in post-vaccination neutralization levels in healthcare workers who experienced a breakthrough infection and matched controls during the Delta wave [38]. The authors found that neutralizing titres were lower against Alpha and Delta variants than the wild-type virus, but were comparable between cases and controls.

Studies reporting associations between antibody levels and VE.

Five studies described correlations between antibody levels and VE against BNT162b2 [30, 31, 33, 34], mRNA-1273 [31–34], ChAdOx1 nCoV-19 [30, 31, 33, 34], Ad26.COV2.S (Janssen/ Johnson and Johnson) [30, 31, 33, 34], NVX-CoV2373 (Novavax) [30, 31, 34], CoronaVac (SinoVac) [31, 34], and rAd26+S+rAd5-S (Gamaleya Research Institute) [31, 34] vaccine using re-analyzed clinical trial and other vaccine. The studies generated correlations using either neutralizing antibody levels, derived through plaque reduction neutralization tests (PRNT) or microneutralization assays, or IgG levels measured through ELISAs.

Three of five studies [30, 31, 34] reported correlation coefficients for the relationship between neutralizing antibodies and VE ranging from 0.79 to 0.96. Two studies [31, 33] reported correlation coefficients of 0.82 to 0.94 to describe the relationship between anti-Spike IgG and VE. Since serology and neutralization assays were not calibrated to a common standard, three studies [30, 31, 34] normalized antibody concentrations against convalescent sera used in their respective clinical trials, and reported antibody concentrations as a ratio of the antibody concentration/convalescent serum concentration. The remaining two studies [32, 33] provided results using the WHO IS.

Using different statistical methods, three studies [31–33] attempted to quantitate the contribution of antibodies to VE measures. Earle et al. incorporated data from seven vaccine clinical trials and reported that neutralizing antibodies accounted for 77.5% to 84.4% of VE [31]. Gilbert et al. focused on mRNA-1273 clinical trial data and reported that neutralizing antibodies accounted for 48.5% (95% CI 34.5, 62.4%) to 68.5% (95% CI 58.5, 78.4%) of VE [32]. This approach was also taken to estimate the effect of anti-S antibodies, with Earle and colleagues finding that anti-S antibody accounts for 91.3% to 94.2% (no CIs provided) of variation in efficacy [31]. Using data from individuals vaccinated with BNT162b2, mRNA-1273, ChAdOx1 nCoV-19 or Ad26.COV2.S, Goldblatt et al. reported that anti-S antibodies account for 68.6% to 97.4% (no CIs provided) of variation in efficacy [33].

Two studies estimated a SARS-CoV-2 threshold of protection. Goldblatt et al. used a random effects meta-analytic approach to calculate protective thresholds in WHO IS units for ancestral strain SARS-CoV-2 and Alpha (B.1.1.7) of 154 (95% CI 42, 559) and 171 (95% CI 57, 519) anti-S binding antibody units (BAU/mL), respectively. Khoury and colleagues used a protective neutralization classification model to estimate the antibody concentration resulting in 100% protection, which they estimated to be 28.6% (95% CI 19.2–29.2%) of the mean convalescent antibody level [34]. The authors also applied a logistic model to calculate the 50% protective neutralization level for symptomatic disease (the titre at which 50% of individuals are protected from symptomatic infection), which was found to be 20.2% (95% CI 14.4, 28.4) of the mean convalescent antibody level. This level corresponded to a neutralization titre of between 1:10 to 1:30 in most assays, which the authors estimate corresponds to 54 (95% CI 30–96) international units (IU)/ml. For severe disease, the 50% threshold was estimated to be only 3% (95% CI 0.71, 13.0%) of the mean convalescent level.

Quality assessment

During quality assessment (S2 Table), we excluded studies that provided inadequate antibody measures or were missing sampling dates, data or laboratory methods details. Of the included studies, we noted that few reported antibody levels at 30–60 days post infection or vaccination or within 30 days of re-infection or breakthrough [20–22, 26, 28, 35, 37], the time periods which would provide the most insight on antibody levels.

Discussion

Our systematic review found mixed evidence regarding a SARS-CoV-2 CoP, with a lack of standardization between laboratory methodology, assay targets, and sampling time points complicating comparisons and interpretation. Studies examining the relationship between antibody levels and VE presented high correlation coefficients, despite utilizing diverse data that included several vaccines and a variety of assays, VE endpoints and populations [30, 31, 33, 34]. The robust correlations despite data heterogeneity support the concept of an anti-S antibody or neutralizing antibody CoP. Furthermore, several studies that explored differences in GMTs between cases and non-cases [29, 32] or associations between antibody levels and viral load with infection incidence or risk [22, 29, 32, 39], found statistically significant differences and associations. Taken together, these aggregate data reports support an antibody target as a potential correlate. However, individual-level data provided contradictory findings. Individuals described in case reports experienced re-infection or breakthrough infection with considerable anti-S or neutralizing antibody levels pre-infection. Studies that attempted to estimate the contribution of antibody levels to VE measures [31–33] found that a substantial proportion of VE was not explained by antibody levels, suggesting that antibodies are only one component of protection. These findings echo SARS-CoV-2 vaccine trial data showing protection after one dose with very low levels of neutralizing antibodies, and suggest that cellular immunity or non-neutralizing antibodies may also play a role in protection [31, 41].

Our review of the literature indicates that a humoral SARS-CoV-2 CoP may be relative, such that antibodies reduce risk of infection but not eliminated it [4]. An analogous example is the influenza 50% protective dose, defined as the antibody concentration at which the risk of infection is reduced by half [3, 42]. Khoury and colleagues provided evidence for a relative correlate in calculating a “50% protective neutralization level” across vaccine studies, and found that lower antibody levels are required to prevent severe disease than to prevent symptomatic infection [34]. Our findings are also in line with real-world observations where SARS-CoV-2 breakthrough cases are often mild or asymptomatic, suggesting that while there is not adequate immunity to prevent infection, there is adequate immunity to prevent symptomatic or severe disease [43, 44]. Furthermore, since mRNA vaccines produce high antibody levels while viral vector vaccines result in robust cellular immunity, it is also possible that the CoP following vaccination may differ by vaccine product [33, 41].

Other data sources that were not eligible for inclusion in our review are supportive of a humoral CoP. Transfer of SARS-CoV-2 convalescent IgG to naïve rhesus macaques was found to be protective [45], and convalescent plasma and monoclonal antibody therapy have been used clinically [46, 47]. Although neither animal model nor passive transfer of immunity mimics the human immune response precisely, these data underscore the importance of humoral immunity for protection against SARS-CoV-2.

There were several limitations to the available literature for this systematic review. We included several case-reports, which generally provide a lower level of evidence and are prone to bias. The included studies used different laboratory assays and heterogeneity in targets. The WHO IS was seldom used, and the diversity of laboratory assays and results precluded a meta-analysis of our data. To overcome the lack of calibration between laboratory assays, some studies normalized results against convalescent sera. However, since the humoral immune response to natural infection varies by age and disease severity [48], this method is not ideal. Most studies did not report which SARS-CoV-2 lineage. With the emergence of Omicron (B.1.1.529), the lack of Omicron-specific serological data prior to re-infection or breakthrough is unfortunate. Evidence based on in vitro neutralization assays suggests that, for immune responses to Omicron in individuals who have already been exposed to Ancestral SARS-CoV-2 antigens (whether through infection or vaccination), an Omicron CoP may be higher than for Ancestral SARS-CoV-2 or other VOCs, due to the reduced effectiveness of Ancestral antibodies for variant spike protein. To that point, Pfizer-BioNTech has reported a 25-fold reduction in neutralization titres against Omicron compared to Ancestral SARS-CoV-2 in individuals vaccinated with two doses of BNT162b2 [49]. Studies from South Africa and Germany report a reduction in neutralization up to 41-fold [50, 51], despite two or three doses of BNT162b2 or mRNA-1273 and previous infection. However, neutralization levels cannot be interpreted with regards to immunity in the absence of a CoP. This issue will be further complicated as the proportion of individuals with an Omicron-specific immune response due to infection, re-infection or breakthrough increases, especially if the clinical serology tools available for diagnostic purposes continue to use Ancestral SARS-CoV-2 antigens. Since a CoP will undoubtedly be variant-specific, continued study in this area is warranted as further variants are detected and vaccination policies evolve in response.

Our review did not examine the role of cellular immunity, which is a limitation because both animal models and human studies have suggested that cellular immunity is likely integral to protection [45]. Furthermore, the included studies focused on systemic immunity, which limits our ability to comment on mucosal antibodies, a known element of SARS-CoV-2 immunity [52]. Only three studies included in our review measured IgA levels in serum [16, 24, 37]. Since circulating IgA cannot be effectively transported into secretions [53], these studies cannot shed light on potential mucosal correlates of protection.

Our findings emphasize that further research into the role of humoral immunity, including non-neutralizing antibody, Fc effector functions and cellular and mucosal immunity is a priority, especially in the context of immune-evading variants like Omicron. The effect of lineage, vaccine product and the endpoint being measured (i.e. infection, symptomatic disease, severe disease) on the CoP are also essential questions. Currently, 40.5% of the world’s population has not been vaccinated against SARS-CoV-2 [54]. The need to approve more vaccines is urgent, but placebo controlled trials have become difficult to perform [33]. A temporary CoP, even if imperfect, would allow us to break through this impasse by performing non-inferiority studies to authorize new vaccine products.

Taken together, our findings suggest that humoral immunity is an integral part of protection against SARS-CoV-2, and that an antibody target is the most likely immune marker for a SARS-CoV-2 CoP.

Supporting information

S1 Table. Full search strategy.

https://doi.org/10.1371/journal.pone.0266852.s001

(PDF)

S2 Table. PRISMA reporting checklist.

https://doi.org/10.1371/journal.pone.0266852.s002

(DOCX)

S3 Table. Quality appraisal of included manuscripts.

https://doi.org/10.1371/journal.pone.0266852.s003

(ZIP)

Citation: Perry J, Osman S, Wright J, Richard-Greenblatt M, Buchan SA, Sadarangani M, et al. (2022) Does a humoral correlate of protection exist for SARS-CoV-2? A systematic review. PLoS ONE 17(4): e0266852. https://doi.org/10.1371/journal.pone.0266852

References

1. Milne G, Hames T, Scotton C, Gent N, Johnsen A, Anderson RM, et al. Does infection with or vaccination against SARS-CoV-2 lead to lasting immunity? Lancet Respir Med. 2021;9(12):1450–66. pmid:34688434

2. Thornburg NJ. Adaptive immunity and SARS-CoV-2 2021 [https://www.cdc.gov/vaccines/acip/meetings/downloads/slides-2021-09-22/03-COVID-Thornburg-508.pdf.

3. Plotkin SA, Orenstein W., Offit P. A, Edwards K.M. Plotkin’s Vaccines, 7th Edition: Elsevier; 2018.

4. Plotkin SA. Vaccines: correlates of vaccine-induced immunity. Clin Infect Dis. 2008;47(3):401–9. pmid:18558875

5. Krammer F. A correlate of protection for SARS-CoV-2 vaccines is urgently needed. Nat Med. 2021;27(7):1147–8. pmid:34239135

6. Hall VG, Ferreira VH, Ku T, Ierullo M, Majchrzak-Kita B, Chaparro C, et al. Randomized Trial of a Third Dose of mRNA-1273 Vaccine in Transplant Recipients. N Engl J Med. 2021;385(13):1244–6. pmid:34379917

7. Yau K, Abe KT, Naimark D, Oliver MJ, Perl J, Leis JA, et al. Evaluation of the SARS-CoV-2 Antibody Response to the BNT162b2 Vaccine in Patients Undergoing Hemodialysis. JAMA Netw Open. 2021;4(9):e2123622. pmid:34473256

8. Sheward D KC, Pankow A, Castro Dopico X, Martin D, Dillner J, Karlsson Hedestam G, et al. Quantification of the neutralization resistance of the Omicron Variant of Concern. https://drive.google.com/file/d/1CuxmNYj5cpIuxWXhjjVmuDqntxXwlfXQ/view; 2021.

9. Davis C, Logan N, Tyson G, Orton R, Harvey WT, Perkins JS, et al. Reduced neutralisation of the Delta (B.1.617.2) SARS-CoV-2 variant of concern following vaccination. PLoS Pathog. 2021;17(12):e1010022. pmid:34855916

10. Edara VV, Norwood C, Floyd K, Lai L, Davis-Gardner ME, Hudson WH, et al. Infection- and vaccine-induced antibody binding and neutralization of the B.1.351 SARS-CoV-2 variant. Cell Host Microbe. 2021;29(4):516–21 e3. pmid:33798491

11. Hu J, Peng P, Cao X, Wu K, Chen J, Wang K, et al. Increased immune escape of the new SARS-CoV-2 variant of concern Omicron. Cell Mol Immunol. 2022. pmid:35017716

12. WebPlotDigitizer [Version 4.5 Released (August 15, 2021):[https://automeris.io/WebPlotDigitizer/.

13. NIH. Study Quality Assessment Tools [https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools.

14. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Int J Surg. 2021;88:105906. pmid:33789826

15. Ali AM, Ali KM, Fatah MH, Tawfeeq HM, Rostam HM. SARS-CoV-2 reinfection in patients negative for immunoglobulin G following recovery from COVID-19. New Microbes New Infect. 2021;43:100926. pmid:34367645

16. Brehm TT, Pfefferle S, von Possel R, Kobbe R, Norz D, Schmiedel S, et al. SARS-CoV-2 Reinfection in a Healthcare Worker Despite the Presence of Detectable Neutralizing Antibodies. Viruses. 2021;13(4).

17. Dimeglio C, Herin F, Miedouge M, Martin-Blondel G, Soulat JM, Izopet J. Protection of healthcare workers against SARS-CoV-2 reinfection. Clin Infect Dis. 2021;73(7):1323–4. pmid:33503228

18. Gallais F, Gantner P, Bruel T, Velay A, Planas D, Wendling M, et al. Anti-SARS-CoV-2 Antibodies Persist for up to 13 Months and Reduce Risk of Reinfection. medRxiv 2021050721256823; https://doiorg/101101/2021050721256823.

19. Inada M, Ishikane M, Terada M, Matsunaga A, Maeda K, Tsuchiya K, et al. Asymptomatic COVID-19 re-infection in a Japanese male by elevated half-maximal inhibitory concentration (IC50) of neutralizing antibodies. J Infect Chemother. 2021;27(7):1063–7. pmid:33962861

20. Krutikov M, Palmer T, Tut G, Fuller C, Shrotri M, Williams H, et al. Incidence of SARS-CoV-2 infection according to baseline antibody status in staff and residents of 100 long-term care facilities (VIVALDI): a prospective cohort study. Lancet Healthy Longev. 2021;2(6):e362–e70. pmid:34104901

21. Leidi A, Koegler F, Dumont R, Dubos R, Zaballa ME, Piumatti G, et al. Risk of reinfection after seroconversion to SARS-CoV-2: A population-based propensity-score matched cohort study. Clin Infect Dis. 2021(May 27):ciab495.

22. Lumley SF, O’Donnell D, Stoesser NE, Matthews PC, Howarth A, Hatch SB, et al. Antibody Status and Incidence of SARS-CoV-2 Infection in Health Care Workers. N Engl J Med. 2021;384(6):533–40. pmid:33369366

23. Roy S. COVID-19 Reinfection in the Face of a Detectable Antibody Titer. Cureus. 2021;13(3):e14033. pmid:33767941

24. Selhorst P, Van Ierssel S, Michiels J, Marien J, Bartholomeeusen K, Dirinck E, et al. Symptomatic SARS-CoV-2 reinfection of a health care worker in a Belgian nosocomial outbreak despite primary neutralizing antibody response. Clin Infect Dis. 2021;73(9):e2985–e91. pmid:33315049

25. Ul-Haq Z, Khan A, Fazid S, Noor F, Yousafzai YM, Sherin A. First documented reinfection of SARS-COV-2 in second wave from Pakistan. J Ayub Med Coll Abbottabad. 2020;32(Suppl 1)(4):S704–S5. pmid:33754536

26. Vetter P, Cordey S, Schibler M, Vieux L, Despres L, Laubscher F, et al. Clinical, virologic and immunologic features of a mild case of SARS-CoV-2 reinfection. Clin Microbiol Infect. 2021;27(5):791.e1–.e4. pmid:33618012

27. Wilkins JT, Hirschhorn LR, Gray EL, Wallia A, Carnethon M, Zembower TR, et al. Serologic Status and SARS-CoV-2 Infection over 6 Months of Follow Up in Healthcare Workers in Chicago: A Cohort Study. Infect Control Hosp Epidemiol. 2021:1–9. pmid:34369331

28. Munivenkatappa A, Sahay RR, Deshpande GR, Patil DY, Shete AM, Sapkal GN, et al. A case with SARS-CoV-2 reinfection from India. Indian J Med Microbiol. 2021. pmid:34789389

29. Bergwerk M, Gonen T, Lustig Y, Amit S, Lipsitch M, Cohen C, et al. Covid-19 Breakthrough Infections in Vaccinated Health Care Workers. N Engl J Med. 2021;385(16):1474–84. pmid:34320281

30. Cromer D, Juno JA, Khoury D, Reynaldi A, Wheatley AK, Kent SJ, et al. Prospects for durable immune control of SARS-CoV-2 and prevention of reinfection. Nat Rev Immunol. 2021;21(6):395–404. pmid:33927374

31. Earle KA, Ambrosino DM, Fiore-Gartland A, Goldblatt D, Gilbert PB, Siber GR, et al. Evidence for antibody as a protective correlate for COVID-19 vaccines. Vaccine. 2021;39(32):4423–8. pmid:34210573

32. Gilbert PB, Montefiori DC, McDermott A, Fong Y, Benkeser DC, Deng W, et al. Immune Correlates Analysis of the mRNA-1273 COVID-19 Vaccine Efficacy Trial. medRxiv. 2021;https://www.medrxiv.org/content/10.1101/2021.08.09.21261290v4. pmid:34401888

33. Goldblatt D F-GA, Johnson M et al. A Population-Based Threshold of Protection for COVID-19 Vaccines. PREPRINT (Version 1) available at Research Square [https://doiorg/1021203/rs3rs-832531/v1].

34. Khoury DS, Cromer D, Reynaldi A, Schlub TE, Wheatley AK, Juno JA, et al. Neutralizing antibody levels are highly predictive of immune protection from symptomatic SARS-CoV-2 infection. Nat Med. 2021;27(7):1205–11. pmid:34002089

35. Michos A, Tatsi EB, Filippatos F, Dellis C, Koukou D, Efthymiou V, et al. Association of total and neutralizing SARS-CoV-2 spike -receptor binding domain antibodies with epidemiological and clinical characteristics after immunization with the 1(st) and 2(nd) doses of the BNT162b2 vaccine. Vaccine. 2021;39(40):5963–7. pmid:34400017

36. Schulte B, Marx B, Korencak M, Emmert D, Aldabbagh S, Eis-Hubinger AM, et al. Case Report: Infection With SARS-CoV-2 in the Presence of High Levels of Vaccine-Induced Neutralizing Antibody Responses. Front Med (Lausanne). 2021;8:704719. pmid:34368197

37. Strafella C, Caputo V, Guerrera G, Termine A, Fabrizio C, Cascella R, et al. Case Report: Sars-CoV-2 Infection in a Vaccinated Individual: Evaluation of the Immunological Profile and Virus Transmission Risk. Front Immunol. 2021;12:708820. pmid:34249017

38. Yamamoto S, Maeda K, Matsuda K, Tanaka A, Horii K, Okudera K, et al. COVID-19 breakthrough infection and post-vaccination neutralizing antibody among healthcare workers in a referral hospital in Tokyo: a case-control matching study. Clin Infect Dis. 2021.

39. Feng S, Phillips DJ, White T, Sayal H, Aley PK, Bibi S, et al. Correlates of protection against symptomatic and asymptomatic SARS-CoV-2 infection. Nat Med. 2021;27(11):2032–40. pmid:34588689

40. World Health Organization. WHO/BS.2020.2403 Establishment of the WHO International Standard and Reference Panel for anti-SARS-CoV-2 antibody [https://www.who.int/publications/m/item/WHO-BS-2020.2403.

41. Liu X, Shaw RH, Stuart ASV, Greenland M, Aley PK, Andrews NJ, et al. Safety and immunogenicity of heterologous versus homologous prime-boost schedules with an adenoviral vectored and mRNA COVID-19 vaccine (Com-COV): a single-blind, randomised, non-inferiority trial. Lancet. 2021;398(10303):856–69. pmid:34370971

42. Hobson D, Curry RL, Beare AS, Ward-Gardner A. The role of serum haemagglutination-inhibiting antibody in protection against challenge infection with influenza A2 and B viruses. J Hyg (Lond). 1972;70(4):767–77. pmid:4509641

43. Mhawish H, Mady A, Alaklobi F, Aletreby W, Asad T, Alodat M, et al. Comparison of severity of immunized versus non-immunized COVID-19 patients admitted to ICU: A prospective observational study. Ann Med Surg (Lond). 2021;71:102951. pmid:34667593

44. Public Health Ontario Enhanced Epidemiological Summary: COVID-19 Infection in Children: January 15, 2020 to June 30, 2021.

45. McMahan K, Yu J, Mercado NB, Loos C, Tostanoski LH, Chandrashekar A, et al. Correlates of protection against SARS-CoV-2 in rhesus macaques. Nature. 2021;590(7847):630–4. pmid:33276369

46. Katz LM. (A Little) Clarity on Convalescent Plasma for Covid-19. N Engl J Med. 2021;384(7):666–8. pmid:33440086

47. Food and Drug Administration. FDA authorizes REGEN-COV monoclonal antibody therapy for post-exposure prophylaxis (prevention) for COVID-19 2021 [https://www.fda.gov/drugs/drug-safety-and-availability/fda-authorizes-regen-cov-monoclonal-antibody-therapy-post-exposure-prophylaxis-prevention-covid-19.

48. Hansen CB, Jarlhelt I, Perez-Alos L, Hummelshoj Landsy L, Loftager M, Rosbjerg A, et al. SARS-CoV-2 Antibody Responses Are Correlated to Disease Severity in COVID-19 Convalescent Individuals. J Immunol. 2021;206(1):109–17. pmid:33208457

49. Pfizer. Pfizer and BioNTech Provide Update on Omicron Variant 2021 [updated Wednesday December 08, 2021. https://www.pfizer.com/news/press-release/press-release-detail/pfizer-and-biontech-provide-update-omicron-variant.

50. Cele S, Jackson L, Khan K, Khoury DS, Moyo-Gwete T, Tegally H, et al. SARS-CoV-2 Omicron has extensive but incomplete escape of Pfizer BNT162b2 elicited neutralization and requires ACE2 for infection. medRxiv. 2021. pmid:34909788

51. Wilhelm AWM, Grikscheit K, Toptan T, Schenk B, Pallas C, Metzler M, et al. Reduced Neutralization of SARS-CoV-2 Omicron Variant by Vaccine Sera and Monoclonal Antibodies. medRxiv. 2021.

52. Rubin EJ, Baden LR, Morrissey S. Audio Interview: Aspects of Covid-19 Immunity. N Engl J Med. 2021;385(8):e31. pmid:34407350

53. Russell MW, Moldoveanu Z, Ogra PL, Mestecky J. Mucosal Immunity in COVID-19: A Neglected but Critical Aspect of SARS-CoV-2 Infection. Front Immunol. 2020;11:611337. pmid:33329607

54. Ritchie H ME, Rodés-Guirao L, Appel C, Giattino C, Ortiz-Ospina E, Hasell J, et al. Coronavirus Pandemic (COVID-19)". Published online at OurWorldInData.org. 2020 [January 18, 2021]. ’https://ourworldindata.org/coronavirus’.

55. Pan American Health Organization. Interim Guidelines for Detecting Cases of Reinfection by SARS-CoV-2. https://www.paho.org/en/documents/interim-guidelines-detecting-cases-reinfection-sars-cov-2 2020 [.

56. Dagan N, Barda N, Kepten E, Miron O, Perchik S, Katz MA, et al. BNT162b2 mRNA Covid-19 Vaccine in a Nationwide Mass Vaccination Setting. N Engl J Med. 2021;384(15):1412–23. pmid:33626250

57. Polack FP, Thomas SJ, Kitchin N, Absalon J, Gurtman A, Lockhart S, et al. Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine. N Engl J Med. 2020;383(27):2603–15. pmid:33301246