All participants were informed about the study in the course of the medical consultation before vaccination, provided written informed consent, and the study was performed according to the principles of the declaration of Helsinki 2013. The Ethics Committee of the Medical University of Innsbruck approved the use of anonymized leftover specimens of COVID-19 patients or vaccinees (ECS1166/2020) and healthy donors (ECS1166/2018) for scientific purposes.

In this study, 89 people were included and divided into two groups: A (Table 1) covers people older than 70 years (n = 45; average age 83 years (70–94 years)) and B (Table 2) includes adults younger than 60 years (n = 40; average age 44 years (22–61 years)). All participants were fully vaccinated with the BNT162b2 vaccine. None of the enrolled participants had a recent, current, or persistent infectious disease or treatment with steroids and/or immunosuppressants. The percentage of females was 51% for people over 70 year old and 48% for persons younger than 60 years, respectively. For aged participants (group A), the average sampling day after the 1st vaccination was 22 (days 20–28) and 37 (days 34–49) after the 2nd dose, respectively. For people of middle age (group B), the average day of sampling after full vaccination was 35 (days 15–78). Detailed information of each individual from the two groups regarding age, sex, time after first and second vaccinations, as well as inclusion in neutralization and T cell analysis is presented in Tables 1 and 2.

The clinical specimen for SARS-CoV-2 (Delta; B.1.617.2) from COVID-19 positive swab (Ethics statement, ECS1166/2020) and SARS-CoV-2 virus (WT) from repositories (BEI Resources, Manassas, VA, USA; CFAR/NIBSC; Nr-52281) was propagated according to the instructions of the manufacturer and used subsequently in neutralization assays.

The LIAISON^® SARS-CoV-2 TrimericS IgG is a CLIA (Chemiluminescent Immunoassay) that detects IgG antibodies reactive with the spike protein (S1/S2 domain). The assay was performed according to the instructions of the manufacturer and gives the arbitrary units per ml (AU/ml) according to the WHO International Standards for the Anti-SARS-CoV-2 immunoglobulin binding activity (NIBSC 20-136).

VeroE6/TMPRSS2 cells (1.2 × 10⁵) were cultured as described previously (10). On the following day, whole serum or plasma samples after the first dose were serial-diluted from 1:4 to 1:512, while samples taken after the second dose were serial-diluted from 1:8 to 1:1,024. Dilutions were incubated with SARS-CoV-2 wild type or variant strains (4 × 10² PFU·ml⁻¹) for 1 h at 37°C. After incubation, VeroE6/TMPRSS2 cells were inoculated with antibody-opsonized SARS-CoV-2 for 1 h at 37°C with 5% CO₂. After incubation, the inoculum was replaced with culture medium containing 1.5% carboxymethylcellulose (Sigma Aldrich). Cells were incubated for 3 days at 37°C and 5% CO₂ before plaque visualization and counting. For this, cells were washed, fixed, and stained using 0.5% (w/v) Crystal Violet solution (Sigma Aldrich). To determine NT50 and NT90 values from neutralization curves, four-parameter nonlinear regression in GraphPad Prism v.9 was used.

The ELISpot assay was performed using precoated human SARS-CoV-2-specific IFNγ ELISPOT kits (AutoImmun Diagnostika). Peripheral blood was collected, PBMCs were isolated and erythrocytes depleted (RBD-Lyse Buffer Life Technologies). PBMCs were counted and resuspended in X-VIVO medium (X-VIVO TM-10 Serum-free Hematopoietic Cell Medium, Lonza).

In brief, a total of 2 × 10⁵ PBMCs were incubated in duplicates with X-VIVO as a negative control, pokeweed mitogen (AutoImmun Diagnostika GmbH) as a positive control, and 15–20mer peptide pools for SARS-CoV-2 and PanCorona for the four non-SARS human coronaviruses 229E, HKU1, NL63, and OC43 (AutoImmun Diagnostika GmbH) as a control of possible cellular cross-reactive responses. After incubation at 37°C and 5% CO₂ for 20 h, plates were washed with washing buffer (AutoImmun Diagnostika GmbH) and stained with the kit-specific reagents according to the protocol of the manufacturer. Spots were counted using an automated AID ELISPOT reader system (AutoImmun Diagnostika GmbH).

The stimulation index (SI) was calculated by dividing the mean spot numbers in the antigen-specific wells with the mean spot numbers of the negative control. A test was assessed as negative with an SI <2 and positive with an SI ≥2.

Statistical analysis was performed using GraphPad Prism v.9. The statistical significance of SARS-CoV-2-specific antibody titers, T-cell responses, and NT50 and NT90 levels between values of the 3 different time points in the group >70 y.o. was determined using the Wilcoxon test (GraphPad Prism). A Mann–Whitney–U test for nonparametric distribution was applied for comparison of immune responses between groups >70 y.o. and <60 y.o. (GraphPad Prism). The correlation of SARS-CoV-2 Spike T cell response vs. anti-SARS−CoV−2 S IgG as well as anti-SARS−CoV−2 S IgG vs. NT50 WT or Delta variant for participants older than 70 years or younger than 60 years was determined using two-tailed non-parametric Spearman correlation analyses (GraphPad Prism).

To monitor SARS-CoV-2 specific T cells in participants older than 70 years (n = 45; Table 1) before vaccination, after first and second vaccination with BNT162b2, in contrast to second-dose vaccinated participants younger than 60 years (n = 20; Table 2), we collected peripheral blood and analyzed SARS-CoV-2-specific T cell response against the SARS-CoV-2 spike protein. The average age of the elderly group was 83 years, in comparison to the younger participants, who were on average 44 years old. Peripheral blood mononuclear cells (PBMCs) were stimulated with peptide pools to assess the functionality of SARS-CoV-2 spike-specific T cell numbers within the various groups. After 20 h, IFNγ spots were counted, and stimulation indices (SI) were calculated and used as a parameter for the analyzed T cell response against SARS-CoV-2. Figure 1A shows COVID-19-naive participants older than 70 years before vaccination (>70 y.o. dose 0), after first (>70 y.o. dose 1) and second vaccination (>70 y.o. dose 2), in contrast to second-dose vaccinated participants younger than 60 years (<60 y.o. dose 2). The median SI for detection of virus-specific T cells was found to be SI = 1.0 (95% CI, 1.0–1.0), SI = 1.5 (95% CI, 1.0–2.0) and SI = 4.3 (95% CI, 1.7–8.7) for individuals in the elderly group before, after the first and after the second dose of BNT162b2, respectively (Figure 1A; Supplementary Table 1 upper panel). These analyses demonstrate the induction spike-specific T cell numbers after the first vaccination and that they significantly increased between the first and second vaccination in participants older than 70 years. The T cell response against SARS-CoV-2 were lower in the younger group of BNT162b2 vaccinees and showed a SI of 3.3 (95% CI, 1.8–6.2; Figure 1A and Supplementary Table 1 upper part). However, there was no age-dependent significant difference detected between second dose-vaccinated young (<60 y.o. dose 2) or elderly (>70 y.o. dose 2) participants (Figure 1A). Figure 1B additionally displays the individual progression of SARS-CoV-2-spike-specific T cells before and after vaccination with BNT162b2 in people >70 years old as a heatmap. The percentages of participants older than 70 years with positive or negative T cell response against SARS-CoV-2 Spike protein before vaccination, after first and second vaccination are summarized in Figure 1C. After the first vaccine dose, 37.2% were identified as positive for SARS-CoV-2-specific T cells, which was enhanced due to the second vaccine dose and reached a value of 60.5% (Figure 1C). In comparison, 70.0% of individuals from the younger group demonstrated virus-specific T cells, which demonstrates a 9.5% increase of individuals with specific T cell immunity compared to aged vaccinees. Nonetheless, 39.5% of individuals from the elderly and 30% from the younger group showed no detectable SARS-CoV-2-specific T cell response after the second vaccine dose (Figure 1D). These results show that when using the mRNA-based vaccine BNT162b2 in elderly vaccinees, the second dose is essential in order to obtain a comparable, but still lower virus-specific T cell response than younger participants.

Using ELISA-based testing methods, we investigated the presence of SARS-CoV-2-specific IgG antibodies against the S region of the SARS-CoV-2 spike protein in serum samples. Figure 2A demonstrates the levels of IgG antibodies against the S region of the SARS-CoV-2 spike protein before vaccination (4.8; 95% CI 4.8–4.8), after the first (142.0; 95% CI 107.0–260.0) and second (1,330.0; 95% CI 1,120.0–1,600.0) vaccine dose of participants >70 years old. The IgG antibodies were dramatically increased in elderly vaccines after the second dose of BNT162.b2 and reached a comparable amount to the younger population (1,512.0; 95% CI 1,291.0–1,827.0), which was twice vaccinated (Figure 2A; Supplementary Table 1, middle part). The individual progression of IgG antibodies before and after vaccination with BNT162b2 in older people is shown in a heatmap in Figure 2B. After the first vaccine dose, 78.8% of the participants >70 years old already had a positive IgG antibody response (Figure 2C), which was further increased to 100% after the second dose and therefore identical to the results of the younger population (100%) (Figure 2D). The results highlight that the IgG antibody response is equivalent between the two age cohorts after the second dose of BNT162b2.

To further investigate the differences after vaccination between the >70-year-old and <60-year-old populations, we determined the virus neutralization abilities of these two groups by analyzing half maximal neutralization titers (NT50, Figure 3) and neutralization titers that inhibit 90% of viral infection (NT90, Supplementary Figure 1). In this regard, serum samples were incubated at various concentrations with SARS-CoV-2 WT. Analysis of the NT50 values of participants >70 years revealed that between the first and the second vaccination, virus neutralization abilities significantly increased (68.1; 95% Cl 37.8–119.6) and resulted in comparable NT50 values between the two age cohorts after the second vaccine (>70 years old: 481.8; 95% Cl 415.7–768; <60 years old: 536; 95% Cl 134.3–909.4) (Figure 3A; Supplementary Table 1, lower part). These results were also confirmed for the NT90 values (Supplementary Figure 1A). The individual progression of NT50 (Figure 3B) and NT90 (Supplementary Figure 1B) values before and after first and second vaccination with BNT162b2 in the elderly vaccinees is shown as a heatmap. We further analyzed the data and found that 66.7% or 28.9% of >70-year-old individuals demonstrated positive NT50 or NT90 values against SARS-CoV-2 WT, respectively (Figure 3C; Supplementary Figure 1C). Positive neutralization of NT50 after the second vaccination was detected in all individuals of both age groups (100%, Figure 3D). Additionally, 88.9% of the aged participants showed positive NT90 neutralization, which was lower than the younger group (93.8%) (Supplementary Figure 1D).

At the time of analysis, the B.1.617.2 (Delta) variant was the dominant circulating variant worldwide, causing increasing severity of disease progression and more transmissions (11, 12). Therefore, we were interested in analyzing neutralization titers in this VOC of SARS-CoV-2. In general, neutralization titers against the Delta variant were impaired in both age cohorts compared to SARS-CoV-2 WT, and median NT50 values after the second vaccine dose decreased from 536.0 to 231.5 and from 481.8 to 172.6 for young and elderly participants, respectively (Figures 4C, D, Supplementary Table 1, lower part). Similar to SARS-CoV-2 WT, NT50 (Figure 4A) and NT90 (Supplementary Figure 2A) titers against SARS-CoV-2 Delta significantly increased between the first and second vaccine doses in participants >70 years old, similar to SARS-CoV-2 WT. After the second dose of BNT162b2, no significant difference in NT50 or NT90 values were detected between the two age cohorts (NT50 of >70 years old: 172.6; 95% Cl 137.2–240.5; NT50 of <60 years old: 231.5; 95% Cl 131.1–309.9; Supplementary Table 1, lower part). The individual progression of NT50 (Figure 4B) and NT90 (Supplementary Figure 2B) values against SARS-CoV-2 Delta before and after the first as well as after the second vaccination in the aged population is displayed as heatmaps. After the first vaccination, 22.2% of the participants >70 years old had positive neutralization titers (NT50) against the SARS-CoV-2 Delta variant (Figure 4C), which was further increased to 95.6% after the second dose and therefore comparable to the results of the younger population (87.4%) (Figure 4D). In contrast to NT50, NT90 was decreased and only 6.7% of the elderly vaccinees after the first vaccination and 75.6% after the second vaccination were positive (Supplementary Figure 2C). In the younger participants, 81.3% were positive after the second dose (Supplementary Figure 2D).

Results from IgG antibody titers, T cell responses, and neutralization assays (NT50 values) observed in double vaccinated people >70 and <60 years old were statistically analyzed to identify a correlation. First, we correlated IgG antibody titers and T cell responses against SARS-CoV-2 spike protein from each individual in the elderly (Figure 5A) and in the younger (Figure 5B) cohorts using two-tailed non-parametric Spearman correlation analyses. For IgG/T cell correlation, we could not detect any significant correlation in either age group. In comparison, the correlation of IgG antibody titers and NT50 values against SARS-CoV-2 WT was significantly positive in aged (r = 0.6325; p <0.0001) (Figure 5C) and young participants (r = 0.5925; p = 0.0004) (Figure 5D). Similar results were also obtained for the calculation of IgG/NT50 against the SARS-CoV-2 Delta variant. In fact, we found a significant positive correlation for >70-year-old people (r = 0.4645; p = 0.0013) (Figure 5E) and for <60-year-old vaccinees (r = 0.58; p = 0.0005) (Figure 5F).

Finally, we compared the humoral immune response and virus neutralization between females and males in >70- and <60-year-old individuals. Virus-specific T cells against SARS-CoV-2 Spike protein (Supplementary Figure 3A) and IgG antibody titers (Supplementary Figure 3B) were not influenced by sex or age. In contrast, NT50 values against SARS-CoV-2 WT significantly differed between females and males in the aged population (Supplementary Figure 3C). We observed increased NT50 values in elderly women (897.5; 95% CI 524.3–958.7) compared with elderly men (522.0; 95% CI 263.7–548.1). Furthermore, we could detect a decrease in NT50 values in <60-year-old women (530.1; 95% CI 167.4–561.2) compared to >70-year-old women. This age- and sex-dependent effect was not present in NT50 values against SARS-CoV-2 Delta (Supplementary Figure 3D).