SARS-CoV-2 antibody immunity across three continents: the West Africa, West Indies, West London Consortium

Abstract

Background: The experience of the COVID-19 pandemic has differed across continents. We hypothesized that regional differences in SARS-CoV-2 immunity might explain this observation. We therefore established the WWW Consortium in Ghana, W Africa; Jamaica, W Indies; and W London. Here, we describe the extent to which antibody immunity differs between these geographic locations.

Methods: The WWW Consortium harmonises across the HERITAGE (Accra, Ghana), WINDFall (Kingston, Jamaica) and Legacy (London, UK) studies, establishing sharing frameworks for samples, metadata, and data; related permissions and oversight; and associated physical and cloud infrastructure. With centralised testing, we performed serological assessments across all three locations at two snapshots in 2024 (April 1^st - August 18^th; August 19^th - December 31^st) using high-throughput live virus neutralization and anti-nucleocapsid IgG, including n=763 individuals.

Findings: We found that across all sites most participants had detectable neutralising antibody titres against JN.1 and XEC - the predominant variants in 2024. There were site-related differences in immunity: vaccine-included SARS-CoV-2 strains were better neutralised by participants from the Legacy study - Ancestral, BA.5, XBB.1.5 initially, and JN.1 after a homologous booster in autumn 2024. For HERITAGE, neutralisation of both alpha- (HCoV-229E) and beta-coronaviruses (HCoV-OC43) was higher than WINDFall suggesting a cross-coronavirus serological response in West Africa. Finally, antigenic cartography identified two distinct antibody landscapes, with JN.1 and XEC antigenically distant in Legacy, but not in HERITAGE and WINDFall.

Interpretation: There is international heterogeneity in SARS-CoV-2 antibody immunity. Global recommendations for vaccine strain selection should incorporate data from diverse populations to ensure accurate, equitable recommendations.

Funding: The Wellcome Trust.

Figure 1 |. The West Africa, West Indies, West London consortium in 2024

(A) Map showing the 3 contributing studies to WWWc. (B) Numbers of daily new confirmed COVID-19 cases per million population over 2024–25 (World Health Organization (2025); Population based on various sources (2024) – with major processing by Our World in Data). Retrieval date: 5 June 2025 (C) The proportion of SARS-CoV-2 sequences corresponding to each of the main variants worldwide. Data from GISAID, with analysis from COVspectrum. Retrieval date: 6 June 2025 (D) Vaccines within WWWc are shown for HERITAGE, WINDFall, and Legacy (E) Serum samples used in this study, splitting 2024 into cohorts Period 1 (April 1^st – August 18^th) and Period 2 (August 19^th – December 31^st). See Tables 1 & 2 for demographic information for Period 1 and Period 2 respectively. Vaccine nomenclature: AZD1222 (ChAdOx-1, Oxford/AstraZeneca; adenoviral vector), BNT162b2 (Pfizer/BioNTech; mRNA vector), mRNA1273 (Moderna; mRNA vector), mRNA1273.214 (Moderna bivalent Ancestral & BA.1; mRNA vector), BNT162b2-XBB.1.5 (Pfizer/BioNTech monovalent XBB.1.5; mRNA vector), BNT162b2-JN.1 (Pfizer/BioNTech monovalent JN.1; mRNA vector), mRNA1273.167 (Moderna monovalent JN.1; mRNA vector).

Figure 2 |. Continental heterogeneity in SARS-CoV-2 serological immunity during 2024

Serum neutralisation titres against SARS-CoV-2 variants from HERITAGE, WINDFall, and Legacy participants at two time points in 2024: (A) Period 1, P1, April 1^st – August 18^th; and (B) Period 2, P2, August 19^th – December 31^st. Titres are compared between time points for JN.1 (C) and XEC (D). Titres are plotted as Log₂-transformed 50% inhibitory concentrations (IC₅₀), the reciprocal of the serum dilution at which 50% of viral infection is inhibited. Significance was determined by two-tailed Wilcoxon rank-sum tests and is indicated by asterisks: * p < 0.05; ** p < 0.01; *** p < 0.001; **** P < 0.0001; ns, not significant (p ≥ 0.05).

Figure 3 |. Spatial differences in antibody immunity in a matched West Africa, West Indies, West London sub-cohort

(A) Serum samples retained after matching, splitting 2024 into time points Period 1 (April 1st – August 18th) and Period 2 (August 19th – December 31st) indicated with the dashed line. See Table 4 & Supplemental Table 1 for demographic information for the matched Period 2 and Period 1, respectively. (B) Serum neutralisation titres in Period 2 2024 (August 19^th – December 31^st) from HERITAGE, WINDFall, and Legacy participants matched based on age group (≤35, 35–50, ≥50), sex, and number of vaccinations (≥2 doses).Titres are plotted as Log₂-transformed 50% inhibitory concentrations (IC₅₀), the reciprocal of the serum dilution at which 50% of viral infection is inhibited. Significance was determined by two-tailed Wilcoxon rank-sum tests and is indicated by asterisks: * p < 0.05; ** p < 0.01; *** p < 0.001; **** P < 0.0001; ns, not significant (p ≥ 0.05).

Figure 4 |. Common coronavirus neutralising immunity differs around the world

Serum neutralisation titres against human coronaviruses OC43 and 229E from HERITAGE, WINDFall, and Legacy participants at two timepoints in 2024: (A) Period 1, April 1^st – August 18^th; and (B) Period 2, August 19^th – December 31^st. Titres are plotted as Log₂-transformed 50% inhibitory concentrations (IC₅₀), the reciprocal of the serum dilution at which 50% of viral infection is inhibited. Significance was determined by two-tailed Wilcoxon rank-sum tests and is indicated by asterisks: * p < 0.05; ** p < 0.01; *** p < 0.001; **** P < 0.0001; ns, not significant (p ≥ 0.05).

Figure 5 |. Geographic differences in antigenic space

Antigenic maps of SARS-CoV-2 variants for each study and timepoint in 2024, with columns HERITAGE, WINDFall, and Legacy (columns left to right) and rows Period 1 (top) and Period 2 (bottom). Coloured circles denote virus antigens; black squares indicate individual sera; the N labels show the count of sera. Maps were generated in Racmacs; one grid unit represents a two-fold change in neutralization titre. Proximity between a serum and a variant reflects stronger neutralisation of that variant.