Impact of SARS-CoV-2 exposure history on antibody kinetics and correlates of protection in The Gambia

BACKGROUNDUnderstanding SARS-CoV-2 antibody dynamics is critical for pandemic preparedness, particularly where population immunity has developed through high infection rates with minimal vaccination. Whether predominantly asymptomatic infections confer protective immunity and which biomarkers best predict protection in resource-limited settings remain unclear. METHODSWe conducted a household cohort study in The Gambia over 15 months (March 2021-June 2022) during Delta and Omicron waves, with weekly upper respiratory tract sampling for SARS-CoV-2 PCR testing. Serum and mucosal samples were collected at baseline, 6, and 12 months. We measured serum neutralising antibodies and mucosal IgA against five SARS-CoV-2 variants. Using Bayesian hierarchical models, we quantified antibody kinetics and evaluated correlates of protection by predictive accuracy across biomarker-variant combinations. FINDINGSAmong 289 participants with 768 serological and mucosal measurements, attack rates of 53% (Delta) and 79% (Omicron) were observed, with 84% of infections asymptomatic. Delta infection produced serum neutralising antibody responses to the Delta variant (27-fold rise at peak, 95% CrI 5{middle dot}0-95{middle dot}2) persisting 197 days >4-fold baseline, while Delta mucosal IgA responses were shorter (9-fold rise at peak, 95% CrI 1{middle dot}7-39{middle dot}3, persisting 62 days). Delta infections generated substantial cross-reactive boosting to Ancestral and Alpha variants, whereas Omicron boosting was more specific to BA.1. Participants with hybrid immunity from infection and vaccination achieved higher antibody levels (24-fold rise) than those with infection alone (6-fold rise) during the Delta wave. Variant-matched neutralisation demonstrated superior predicted protection performance against infection compared to other biomarkers in both waves (Delta: AUC 0{middle dot}66; Omicron BA.1: AUC 0{middle dot}65). INTERPRETATIONPredominantly asymptomatic SARS-CoV-2 infections in The Gambia generated robust, durable and protective antibody responses. Variant-matched serum neutralising antibody levels capture protection as effectively as complex multi-biomarker approaches, providing pragmatic guidance for serological surveillance for population immunity in resource-constrained settings. FUNDINGUnited Kingdom Research and Innovation Grant (No. MC_PC_19084). RESEARCH IN CONTEXTO_ST_ABSEvidence before this studyC_ST_ABSWe searched PubMed Central for studies up to 29th December 2025 using the terms ("SARS-CoV-2" OR "COVID-19") AND "longitudinal serology" AND "sub-Saharan Africa", with no language restrictions. This search yielded four results, of which only one was a longitudinal serological study conducted in sub-Saharan Africa that examined antibody dynamics and reinfection patterns in Kenya. The remaining three were a systematic review and meta-analysis on COVID-19 burden in developing countries, and two serological studies on non-COVID pathogens. In contrast, broader search term ("SARS-CoV-2" OR "COVID-19") AND "longitudinal serology" returned 74 results, highlighting the stark underrepresentation of sub-Saharan African populations in longitudinal serology studies. This gap is critical because sub-Saharan African populations present distinct epidemiological features, including high reported rates of asymptomatic infections (65-85% compared to 20-35% globally), and population immunity predominantly driven by natural infection rather than vaccination. Further, most studies on correlates of protection have focused on vaccinated populations in high-income countries, using clinical trials or standard observational cohorts. Few have employed household-based design with intensive surveillance, which uniquely enables the identification of definite exposure events and precise temporal linkage between exposure and antibody measurements--overcoming a key limitation in which exposure timing is uncertain. Added value of this studyThis household-based cohort study provides high-resolution longitudinal immunological data from The Gambia across 15 months spanning Delta and Omicron waves, including 768 serological and mucosal measurements from 289 participants with intensive weekly PCR surveillance. We measured both serum neutralising antibodies and mucosal IgA against five SARS-CoV-2 variants concurrently, including both Ancestral and Alpha variants and contemporaneous circulating variants (Delta, Omicron BA.1, BA.2). The household design with defined exposure events and temporal antibody measurements provides a more consistent assessment of protection at the time of exposure compared to standard cohort studies. Using Bayesian hierarchical modelling, we quantified antibody kinetics following predominantly asymptomatic natural infections and identified optimal correlates of protection. We demonstrate that natural infection generates robust and durable antibody responses despite minimal symptoms, with Delta infection producing 18-fold rises in neutralising antibodies persisting 101 days above four-fold baseline. Importantly, we show that hybrid immunity, infection followed by vaccination, produces substantially higher antibody responses than infection alone, with 24-fold rises compared to 6-fold rises during the Delta wave. This hybrid immunity advantage was consistently maintained across subsequent exposures, demonstrating that vaccination provides meaningful immunological benefit even in highly infection-experienced populations. We show that prior infection history substantially influences antibody responses to antigenically similar variants through immune imprinting, whereas responses to the antigenically divergent Omicron variant were less affected by infection history. Critically, variant-matched serum pseudoneutralisation provided the correlate of protection with the best predictive capacity compared to other biomarkers and variant combinations, with Delta pseudoneutralisation optimal for the Delta wave (AUC 0{middle dot}66) and Omicron BA.1 pseudoneutralisation optimal for the Omicron wave (AUC 0{middle dot}65). Single functional serum neutralisation assays captured protection as effectively as more complex dual biomarker approaches combining serum and mucosal measurements. Implications of all the available evidenceOur findings demonstrate that populations in sub-Saharan Africa with high natural infection rates and low vaccine coverage develop robust, durable immunity that provides baseline protection against future SARS-CoV-2 waves, even when infections are predominantly asymptomatic. Critically, however, the addition of vaccination to natural immunity produces substantially higher neutralising antibody levels, the optimal correlate of protection identified in this study, indicating that vaccination provides meaningful benefit even in highly seroprevalent populations with extensive prior infection exposure. The household-based design with intensive surveillance provides correlates of protection estimates that may be particularly relevant for assessing protection at defined exposure timepoints in community settings. The finding that single functional neutralisation assays, particularly those measuring variant-specific responses, perform comparably to complex multi-biomarker approaches provides pragmatic guidance for serological surveillance in resource-constrained settings, supporting investment in laboratory capacity for these assays while avoiding unnecessary complexity. The non-linear relationship between antibody levels and protection suggests that booster vaccination strategies can provide meaningful benefit even in highly seroprevalent populations. These insights from an underrepresented region inform evidence-based public health strategies and underscore the importance of conducting immunological research in diverse global contexts to achieve health equity and ensure pandemic preparedness.