T-cell recognition of SARS-CoV-2 antigen peptides following booster vaccination with mRNA and viral-vectored COVID-19 vaccines

Abstract

Durability and breadth of SARS-CoV-2-specific T-cell responses following booster vaccination remain key to guiding next-generation vaccine design, especially in the context of waning antibodies, emerging variants, and diverse immunological backgrounds. This longitudinal study assessed IFN-γ-producing T-cell responses in 46 fully vaccinated adults who received a booster dose of either the BNT162b2 (mRNA) or Ad26.COV2.S (adenoviral vector) vaccine. All participants were nucleocapsid-seropositive at baseline, with antibody levels stable throughout follow-up, indicating prior natural infection and no incident infection during the study period. Using an ex vivo IFN-γ ELISpot assay, responses were measured at four timepoints (baseline, Months 3, 6 and 9) against spike (S; n = 46) and nucleocapsid (N; n = 16) peptides selected by NetMHCpan-4.1 prediction from overlapping 15-mer libraries spanning both proteins. Each peptide was tested individually, allowing every response to be assigned to a defined sequence. Longitudinal change was modelled using linear mixed-effects models, comparing total with median per-peptide responses to distinguish loss of specificities from loss of per-epitope strength. Analyses were stratified by booster type and vaccination strategy (homologous vs. heterologous). Time strongly affected total S-specific IFN-γ responses (F(3, 96.6) = 33.16, p = 8 × 10 −15 ) but not median per-peptide responses (F(3, 95.8) = 0.62, p = 0.606), indicating that the decline reflected contraction in the number of epitopes recognized rather than weakening of individual responses; the same dissociation held within both booster groups. Ad26.COV2.S boosting was associated with broader S-specific epitope recognition and greater durability, with a subset of responses persisting through Month 9, whereas BNT162b2 recipients showed higher cumulative IFN-γ magnitude through Month 6 — a difference already present before boosting (p = 0.012). N-specific responses declined despite stable anti-nucleocapsid antibody, although two N-derived regions (YGANKDGIIW, VTPSGTWLTY) were recognized at every timepoint. Booster platform and vaccination strategy shaped the magnitude, breadth and durability of SARS-CoV-2-specific T-cell responses, with the longitudinal decline appearing to be driven predominantly by narrowing of epitope breadth among Month 3-reactive peptides rather than reduced per-epitope function. The persistently immunogenic S and N epitopes identified here warrant further evaluation as candidates for multi-epitope T-cell vaccine design.