Development of a T cell-based immunodiagnostic system to effectively distinguish SARS-CoV-2 infection and COVID-19 vaccination status

Abstract

Both SARS-CoV-2 infections and COVID-19 vaccines elicit memory T cell responses. Here, we report the development of 2 pools of experimentally defined SARS-CoV-2 T cell epitopes that, in combination with spike, were used to discriminate 4 groups of subjects with different SARS-CoV-2 infection and COVID-19 vaccine status. The overall T cell-based classification accuracy was 89.2% and 88.5% in the experimental and validation cohorts. This scheme was applicable to different mRNA vaccines and different lengths of time post infection/post vaccination and yielded increased accuracy when compared to serological readouts. T cell responses from breakthrough infections were also studied and effectively segregated from vaccine responses, with a combined performance of 86.6% across all 239 subjects from the 5 groups. We anticipate that a T cell-based immunodiagnostic scheme to classify subjects based on their vaccination and natural infection history will be an important tool for longitudinal monitoring of vaccinations and for establishing SARS-CoV-2 correlates of protection.

Keywords: COVID-19; SARS-CoV-2; T cells; breakthrough infection; epitope; immunodiagnostic tool; vaccination; viruses.

Graphical abstract

Figure 1

SARS-CoV-2-specific CD4+ and CD8+ T cell responses in the study groups SARS-CoV-2-specific CD4+ and CD8+ T cell responses were measured as percentages of AIM+ (OX40+CD137+) CD4+ T cells (A and B) or AIM+ (CD69+CD137+) CD8+ T cells (C and D) after stimulation of PBMCs with peptides pools encompassing spike-only (Spike) MPs or the experimentally defined CD4RE and CD8RE MPs representing all the proteome without spike. EVB MP was used as a control. Graphs show individual response of spike, CD4RE, or CD8RE and the combination of both (total CD4+ or total CD8+) plotted as background subtracted (A and C) or as SI (B and D) against DMSO negative control. Geometric mean with standard deviation (SD) for the 4 different groups is shown. Kruskal-Wallis test adjusted with Dunn’s test for multiple comparisons was performed, and p values < 0.05 considered statistically significant. I−V−, unexposed and unvaccinated (n = 30); I+V−, infected and non-vaccinated (n = 30); I+V+, infected and then vaccinated (n = 30); I−V+, non-infected and vaccinated (n = 30). Threshold of positivity (TP) is indicated. Median response and the number or percentage of positive responding donors for each group is shown.

Figure 2

COVID-19 clinical classification scheme using SARS-CoV-2-specific CD4+ T cell responses CD4+ T cell responses to spike and CD4RE MPs were measured as percentage of AIM+ (OX40+CD137+) CD4+ T cells and plotted in 2 dimensions as absolute magnitude in order to discriminate the 4 study groups with known COVID-19 status of infection and/or vaccination in 2 independent cohorts: (A) exploratory cohort (n = 120) and (B) validation cohort (n = 96). I−V−, unexposed and unvaccinated (n = 30 and n = 20); I+V−, infected and non-vaccinated (n = 30 and n = 20); I+V+, infected and then vaccinated (n = 30 and n = 20); I−V+, non-infected and vaccinated (n = 30 and n = 36). Red dotted lines indicate specific cutoffs. Table inserts depict the diagnostic exam results in 4×4 matrix. Sensitivity, specificity, PPV, NPV, and overall percentage of subjects classified correctly is shown.

Figure 3

COVID-19 clinical classification scheme is applicable to different mRNA vaccines and different lengths of time post infection/post vaccination CD4+ T cell responses to spike and CD4RE MPs were measured as percentage of AIM+ (OX40+CD137+) CD4+ T cells and plotted in 2 dimensions as absolute magnitude in order to discriminated between (A) different types of mRNA vaccines (Moderna versus Pfzier) among vaccinated groups (I−V+ and I+V+), (B) different lengths of time post infection among infected groups (I+V− and I+V+), and (C) different lengths of time post vaccination among vaccinated groups (I−V+ and I+V+). Early infection, PSO ≤ 180; late infection, PSO > 180; early post vaccination, PVD ≤ 30; late post vaccination, PVD > 30. I−V+, non-infected and vaccinated (n = 66); I+V−, infected and non-vaccinated (n = 50); I+V+, infected and then vaccinated (n = 50). Red dotted lines indicate specific cutoffs. Table inserts depict the overall percentage of subjects classified correctly.

Figure 4

SARS-CoV-2 T cell and antibody response in breakthrough infection cases: comparison to other study groups (A and B) SARS-CoV-2-specific T cell responses were measured as percentage of (A) AIM+ (OX40+CD137+) CD4+ T cells or (B) AIM+ (CD69+CD137+) CD8+ T cells after stimulation of PBMCs with spike and CD4RE or CD8RE peptide pools. (C) Comparison of anti-spike RBD IgG titers in the plasma of the different study groups. For both T cell and antibody determinations, only donors matching the V+I+ intervals of vaccination and infection (55–271 and 18–93 days, respectively) were plotted. (A-C) Graph bars show geometric mean with SD. Threshold of positivity (TP), median response, and the number or percentage of positive responding donors for each group is indicated. Kruskal-Wallis test adjusted with Dunn’s test for multiple comparisons was performed, and p values < 0.05 considered statistically significant. (D) V+I+ CD4+ T cell responses plotted using the two-dimensional classification scheme with the specific cutoffs attributed to the different study groups (red dotted lines). I−V−, unexposed and unvaccinated (n = 50); I+V−, infected and non-vaccinated (n = 50); I+V+, infected and then vaccinated (n = 50); I−V+, non-infected and vaccinated (n = 66); V+I+, vaccinated and then infected (n = 23).

Figure 5

Overall COVID-19 clinical classification scheme (A and B) CD4+ T cell responses to spike and CD4RE MPs were measured as percentage of AIM+ (OX40+CD137+) CD4+ T cells and plotted in two dimensions as (A) SFCs per million PBMCs or (B) stimulation index (SI), in order to discriminate the 5 study groups with known COVID-19 status of infection and/or vaccination. I−V−, unexposed and unvaccinated (n = 50); I+V−, infected and non-vaccinated (n = 50); I−V+, non-infected and vaccinated (n = 66); I+V+/V+I+, infected and then vaccinated (I+V+, n = 50) merged with vaccinated and then infected (V+I+, n = 23). Red dotted lines indicate specific cutoffs. Table inserts depict the diagnostic exam results in 4×4 matrix. Sensitivity, specificity, PPV, NPV of all the subjects that participated in this study (n = 239), and overall percentage classified correctly is shown.