Among the authorized vaccines in Europe, the vector-based vaccine ChAdOx1 nCoV-19 and the mRNA-vaccines BNT162b2 and mRNA-1273 are given as dual-dose regimens, whereas the vector-vaccine Ad26.COV.2 is licensed as a single dose only. In general, mRNA-vaccines were shown to have a higher efficacy compared to vector-based regimens, and breakthrough infections are most frequently observed after single-dose Ad25.COV-2 administration1–3. This led German and other national authorities to recommend heterologous booster vaccinations for Ad26.COV.2 vaccinated individuals3, although data on immunogenicity and reactogenicity are lacking. As heterologous mRNA-vaccine boosting after priming with the ChAdOx1-nCoV-19 vector vaccine led to an equivalent or superior antibody- and T-cell response as compared to homologous regimens4–7, we hypothesized that heterologous mRNA-boosting may also be able to optimize specific immunity following Ad25.COV-2-priming.
We analyzed immunogenicity and reactogenicity of fifteen individuals who received heterologous BNT162b2-boosting 16.1 [13.0-19.3] weeks after Ad26.COV.2 vaccination. Blood-counts, vaccine-induced IgG, neutralizing activity using a surrogate assay, spike-specific CD4 and CD8 T cells and polyclonally activated T cells were quantified exactly as previously described4. Reactogenicity was self-reported within the first seven days after the first and the second vaccination. Eleven males and four females with a mean age of 29.3±8.6 years were included. Heparinized blood samples for immunological testing were obtained 21 (IQR 21-28) days after the first and 16 (IQR 14-23) days after the booster-vaccination.
As shown in figure 1A and Table 1, a single dose of Ad26.COV.2 led to a poor induction of spike-specific IgG (62.16 [46.87-112.2] BAU/ml). Accordingly, median neutralizing activity was low (29 [12.7-47.9] % inhibition) with positive results in only 2/15 individuals. Heterologous boosting with BNT162b2 led to a significant increase in median IgG-levels (3168 [1896-4985] BAU/ml, 39-fold, p<0.0001) and in neutralizing activity (median 100 [100-100] %, 8-fold, p<0.0001). Spike-specific CD4 and CD8 T cells were stimulated with overlapping peptides and identified based on upregulation of CD69 and induction of IFNγ. As shown in Figure 1B, spike-specific CD4 T-cell levels after Ad26.COV.2 priming were overall low (0.032 [0.019-0.051] %) with only 8/15 individuals above detection limit. In contrast, median levels of spike-specific CD8 T cells were higher (0.405 [0.120-1.038] %). Although the booster effect on CD4 T cells was less pronounced (3-fold) than on CD8 T cells (7-fold), both T-cell subtypes showed a significant increase reaching median levels of 0.070 [0.049-0.157] % CD4 T cells and 1.077 [0.374-3.811] % CD8 T cells (p=0.003 and p=0.002, respectively, Figure 1B). All effects were spike-specific, as boosting had no effect on SEB-reactive CD4 or CD8 T-cell levels (Figure 1C). The heterologous vaccine regimen was well tolerated, and adverse events were reported more pronounced after the Ad26.COV.2 dose (figure 2A and B). Overall, both local and systemic adverse events were rather moderate with largely no difference between the two doses (Figure 2C and D). Among systemic adverse events, chills were the only events which were significantly more frequent after the first Ad26.COV.2 vaccination (figure 2D).
Table 1
Differential blood counts and immunological parameters after the first and the second vaccination
| vaccine | | Ad26.COV.2 first vaccine | BNT162b2 booster vaccine | p-value* |
| Differential blood-counts | | | |
| Leukocytes (cells/µl) | median [IQR] | 5200 [4600-7500] | 5800 [4900-7100] | p=0.552 |
| Lymphocytes (cells/µl) | median [IQR] | 2012 [1706-2446] | 2011 [1835-2361] | p=0.454 |
| Monocytes (cells/µl) | median [IQR] | 470 [419-643] | 538 [446-676] | p=0.303 |
| Granulocytes (cells/µl) | median [IQR] | 2580 [2106-4456] | 3131 [2583-3593] | p=0.345 |
| Thrombocytes (cells/µl) | median [IQR] | 241000 [207000-282000] | 273000 [205000-293000] | p=0.192 |
| Immunological parameters | | | |
| IgG (BAU/ml) | median [IQR] | 62 [47-112] | 3168 [1896-4985] | p<0.0001 |
| % positive§ | 93% [14/15] | 100% [15/15] | p>0.999# |
| Neutralizing activity (% inhibition) | median [IQR] | 29.5 [12.7-47.9] | 100.2 [100.0-100.6] | p<0.0001 |
| % positive§ | 13% [2/15] | 100% [15/15] | p<0.0001# |
| Spike-specific CD4 T cells (%) | median [IQR] | 0.032 [0.019-0.051] | 0.070 [0.049-0.157] | p=0.003 |
| % positive§ | 53% [8/15] | 93% [14/15] | p=0.035# |
| Spike-specific CD8 T cells (%) | median [IQR] | 0.405 [0.120-1.038] | 1.077 [0.374-3.811] | p=0.002 |
| % positive§ | 87% [13/15] | 100% [15/15] | p=0.483# |
| SEB-reactive CD4 T cells (%) | median [IQR] | 4.834 [2.491-6.717] | 4.179 [2.159-6.872] | p=0.934 |
| SEB-reactive CD8 T cells (%) | median [IQR] | 7.885 [5.116-13.22] | 5.928 [9.004-10.420] | p=0.188 |
| *p-values were calculated using the Wilcoxon matched-pairs test or #Fisher´s exact test. §refers to the detection limits (see methods). |
Despite investigated in a small sample size, a heterologous boost with an mRNA vaccine led to a substantial increase in both humoral and cellular immunity in all tested individuals. The immunogenicity profile after vaccination with the Ad26.COV.2 vector shares similarities with primary vaccination with the ChAdOx1 nCoV-19 vector where spike-specific T cells were more strongly induced as compared to antibodies8. The increase after heterologous boosting is in a similar range as previously described for individuals after homologous BNT162b2 or heterologous ChAdOx1 nCoV-19/BNT162b2-vaccination4. Thus, together with evidence of a less profound booster-effect after homologous ChAdOx1 nCoV-19 vector-vaccination4,7 and recent data on high effectiveness of heterologous ChAdOx1 nCoV-19/mRNA vaccination9,10, our data support current recommendations for heterologous booster-vaccination to optimize specific immunity and protection against SARS-CoV-2 infection in Ad26.COV.2-vaccinated individuals. Given the moderate spectrum of adverse events, heterologous boosting appears safe and may raise confidence in following recommendations towards heterologous vaccination.