Persistence of spike-specific immune responses in BNT162b2-vaccinated donors and generation of rapid ex-vivo T cells expansion protocol for adoptive immunotherapy: A pilot study

Abstract

Introduction: The BNT162b2 mRNA-based vaccine has shown high efficacy in preventing COVID-19 infection but there are limited data on the types and persistence of the humoral and T cell responses to such a vaccine.

Methods: Here, we dissect the vaccine-induced humoral and cellular responses in a cohort of six healthy recipients of two doses of this vaccine.

Results and discussion: Overall, there was heterogeneity in the spike-specific humoral and cellular responses among vaccinated individuals. Interestingly, we demonstrated that anti-spike antibody levels detected by a novel simple automated assay (Jess) were strongly correlated (r=0.863, P<0.0001) with neutralizing activity; thus, providing a potential surrogate for neutralizing cell-based assays. The spike-specific T cell response was measured with a newly modified T-spot assay in which the high-homology peptide-sequences cross-reactive with other coronaviruses were removed. This response was induced in 4/6 participants after the first dose, and all six participants after the second dose, and remained detectable in 4/6 participants five months post-vaccination. We have also shown for the first time, that BNT162b2 vaccine enhanced T cell responses also against known human common viruses. In addition, we demonstrated the efficacy of a rapid ex-vivo T cell expansion protocol for spike-specific T cell expansion to be potentially used for adoptive-cell therapy in severe COVID-19, immunocompromised individuals, and other high-risk groups. There was a 9 to 13.7-fold increase in the number of expanded T cells with a significant increase of anti-spike specific response showing higher frequencies of both activation and cytotoxic markers. Interestingly, effector memory T cells were dominant in all four participants' CD8+ expanded memory T cells; CD4+ T cells were dominated by effector memory in 2/4 participants and by central memory in the remaining two participants. Moreover, we found that high frequencies of CD4+ terminally differentiated memory T cells were associated with a greater reduction of spike-specific activated CD4+ T cells. Finally, we showed that participants who had a CD4+ central memory T cell dominance expressed a high CD69 activation marker in the CD4+ activated T cells.

Keywords: COVID-19 vaccine; SARS-CoV-2; spike-specific T cells expansion; spike-specific immune responses; surrogate neutralization.

Figure 1

Interpersonal variation of anti-spike binding and neutralizing antibodies responses over time among BNT162b2 vaccinated participants. Serum samples were collected from six BNT162b2 vaccinated participants at the baseline, 20-, 34-, and 150-days post-vaccination. The anti-S and anti-S1 IgG levels were assessed using Jess and ELISA, respectively. The anti-S neutralizing activity was measured using the neutralization assay. (A) Anti-S IgG response in six BNT162b2 vaccinated participant over time (Jess). (B) Anti-S1 IgG response in six BNT162b2 vaccinated participant over time (ELISA). (C) Anti-S neutralizing activity in each BNT162b2 vaccinated participant over time (Neutralization assay). The lines indicate the cut-off value of a positive antibody response.

Figure 2

Correlation between the anti-spike binding and neutralizing antibodies responses induced by BNT162b2 vaccination and detected by three different immunoassays. Serum samples were collected from six BNT162b2 vaccinated participants at the baseline, 20-, 34-, and 150-days post-vaccination. The anti-S and anti-S1 IgG levels were assessed using Jess and ELISA, respectively. The anti-S neutralizing activity was measured using the neutralization assay. (A–C) Heat-map of anti-S IgG, anti-S1 IgG, and anti-S neutralizing activity responses in six BNT162b2 vaccinated participants over time. (D) Dynamics of the anti-S IgG levels in six BNT162b2 vaccinated participants over time (Jess). (E) Dynamics of the anti-S neutralizing activity in six BNT162b2 vaccinated participants (Neutralization assay). Each symbol represents an individual participant with a line indicating the median of each time point. One-way ANOVA test was used, P value was considered statistically significant when *P ≤ 0.05. All samples were run in duplicates. (F–I) Correlation between anti-S, S1 RBD, S2, and S1 IgG levels detected by JESS and neutralizing activity, respectively. (J) Correlation between anti-S1 IgG levels detected by semi-quantitative ELISA and neutralizing activity. (K) Correlation between anti-S1 IgG levels detected by ELISA and anti-S IgG levels detected by JESS. All correlations were analyzed by Pearson statistical test and linear regression models. The scatter point represents serum samples (n=24) collected from six BNT162b2 vaccinated participants at the baseline, 20-, 34-, and 150-days post-vaccination, and the blue error band represents the 95% confidence interval. The coefficient of correlation (r) represents the strength of the linear relationship between the different immunoassays. The coefficient of determination (R2) represents the percentage of variance in the given data set. The P-value tests whether the regression equation is significant. P value was considered statistically significant when *P ≤ 0.05. The stars present the level of significance. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 3

Spike-specific and spike cross-reactive T cell responses in BNT162b2 vaccinated participants. (A) T cell response to S1 peptide pools in six BNT162b2 vaccinated participant at the baseline, 20-, 34-, and 150-days post-vaccination. (B) Heat-map of S1- specific T cell responses in six BNT162b2 vaccinated participants over time (C) Dynamics of S1- specific T cell responses in six BNT162b2 vaccinated participants over time. Each symbol represents an individual participant with a line indicating the median of each time point. (D) T cell response to high homology peptide pools in BNT162b2 vaccinated participants at baseline (Day 0) and on day 34 post-vaccination (after the second dose) (E) T cell response to five human common viruses peptides (CMV, EBV, AdV 3 and 5 and BKV) in BNT162b2 vaccinated participants at baseline (Day 0) and on day 34 post-vaccination (after the second dose). Results are presented as the mean of the number of spots forming cells (SFCs) per 250 000 PBMCs subtracting the background (negative control) count. A positive response was defined as an SFCs of 10 or more. One-way ANOVA test was used, P value was considered statistically significant when *P ≤ 0.05. All samples were run in duplicates. The stars present the level of significance. *P < 0.05; **P < 0.01.

Figure 4

Expansion and functional characterization of expanded spike-specific T cells. (A) IFN-γ secretion by pre-expansion PBMCs collected 150 days post-vaccination and spike-specific expanded T cells following S1 peptide pools overnight stimulation. Results are presented as the mean of the number of spots forming cells (SFCs) per 250 000 PBMCs subtracting the background (negative control) count. A positive response was defined as an SFCs of 10 or more. Student t test was used, P value was considered statistically significant when *P ≤ 0.05. All samples were run in duplicates. (B) Viable cell counts, and fold expansion were assessed on days 0, 6, and 11 of expansion using trypan blue exclusion dye. (C) Correlation between IFN-γ response folds increase and the proliferative expansion folds increase between pre-expansion PBMCs and spike-specific expanded T cells in four BNT162b2 vaccinated participants. The dotted lines represent confidence intervals at 95%. The stars present the level of significance. *P < 0.05; **P < 0.01; ****P < 0.0001.

Figure 5

Phenotyping of the spike-specific expanded T cells. (A) Frequency of CD4⁺ OX40⁺ CD69⁺ and CD8⁺ CD137⁺ CD69⁺ activated T cells within pre-expansion and expanded T cells following S1 stimulation in four BNT162b2 vaccinated participants. (B) Frequency of CD3⁺ CD107⁺, CD4⁺ CD107⁺, and CD8⁺ CD107⁺ cytotoxic T cells within pre-expansion and expanded T cells following S1 stimulation in four BNT162b2 vaccinated participants. (C) Frequency of CD4⁺ naïve (CCR7⁺CD45RA⁺), central memory (CCR7⁺CD45RA⁻), effector memory (CCR7⁻CD45RA⁻), and terminally differentiated memory (CCR7⁻CD45RA⁺) within pre-expansion and expanded T cells following S1 stimulation in four BNT162b2 vaccinated participants. (D) Frequency of CD8⁺ naïve (CCR7⁺CD45RA⁺), central memory (CCR7⁺CD45RA⁻), effector memory (CCR7⁻CD45RA⁻), and terminally differentiated memory (CCR7⁻CD45RA⁺) within pre-expansion and expanded T cells following S1 stimulation in four BNT162b2 vaccinated participants.