Antibody response to SARS-CoV-2 infection in humans: A systematic review

• Published in: PLoS ONE 2020, Vol.15 (12), p.e0244126
• Main Authors: Post, Nathan, Eddy, Danielle, Huntley, Catherine, van Schalkwyk, May C. I, Shrotri, Madhumita, Leeman, David, Rigby, Samuel, Williams, Sarah V, Bermingham, William H, Kellam, Paul, Maher, John, Shields, Adrian M, Amirthalingam, Gayatri, Peacock, Sharon J, Ismail, Sharif A, Mantis, Nicholas J
• Format: Report
• Language: English
• Subjects: Antibodies; Health aspects; Viral antibodies
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0244126

Background Progress in characterising the humoral immune response to Severe Acute Respiratory Syndrome 2 (SARS-CoV-2) has been rapid but areas of uncertainty persist. Assessment of the full range of evidence generated to date to understand the characteristics of the antibody response, its dynamics over time, its determinants and the immunity it confers will have a range of clinical and policy implications for this novel pathogen. This review comprehensively evaluated evidence describing the antibody response to SARS-CoV-2 published from 01/01/2020-26/06/2020. Methods Systematic review. Keyword-structured searches were carried out in MEDLINE, Embase and COVID-19 Primer. Articles were independently screened on title, abstract and full text by two researchers, with arbitration of disagreements. Data were double-extracted into a pre-designed template, and studies critically appraised using a modified version of the Public Health Ontario Meta-tool for Quality Appraisal of Public Health Evidence (MetaQAT) tool, with resolution of disagreements by consensus. Findings were narratively synthesised. Results 150 papers were included. Most studies (113 or 75%) were observational in design, were based wholly or primarily on data from hospitalised patients (108, 72%) and had important methodological limitations. Few considered mild or asymptomatic infection. Antibody dynamics were well described in the acute phase, up to around three months from disease onset, but the picture regarding correlates of the antibody response was inconsistent. IgM was consistently detected before IgG in included studies, peaking at weeks two to five and declining over a further three to five weeks post-symptom onset depending on the patient group; IgG peaked around weeks three to seven post-symptom onset then plateaued, generally persisting for at least eight weeks. Neutralising antibodies were detectable within seven to 15 days following disease onset, with levels increasing until days 14-22 before levelling and then decreasing, but titres were lower in those with asymptomatic or clinically mild disease. Specific and potent neutralising antibodies have been isolated from convalescent plasma. Cross-reactivity but limited cross-neutralisation with other human coronaviridae was reported. Evidence for protective immunity in vivo was limited to small, short-term animal studies, showing promising initial results in the immediate recovery phase. Conclusions Literature on antibody responses to SARS