Ad26.COV2.S protects Syrian hamsters against G614 spike variant SARS-CoV-2 and does not enhance respiratory disease

Abstract

Previously we have shown that a single dose of recombinant adenovirus serotype 26 (Ad26) vaccine expressing a prefusion stabilized SARS-CoV-2 spike antigen (Ad26.COV2.S) is immunogenic and provides protection in Syrian hamster and non-human primate SARS-CoV-2 infection models. Here, we investigated the immunogenicity, protective efficacy, and potential for vaccine-associated enhanced respiratory disease (VAERD) mediated by Ad26.COV2.S in a moderate disease Syrian hamster challenge model, using the currently most prevalent G614 spike SARS-CoV-2 variant. Vaccine doses of 1 × 10⁹ and 1 × 10¹⁰ VP elicited substantial neutralizing antibodies titers and completely protected over 80% of SARS-CoV-2 inoculated Syrian hamsters from lung infection and pneumonia but not upper respiratory tract infection. A second vaccine dose further increased neutralizing antibody titers that was associated with decreased infectious viral load in the upper respiratory tract after SARS-CoV-2 challenge. Suboptimal non-protective immune responses elicited by low-dose A26.COV2.S vaccination did not exacerbate respiratory disease in SARS-CoV-2-inoculated Syrian hamsters with breakthrough infection. In addition, dosing down the vaccine allowed to establish that binding and neutralizing antibody titers correlate with lower respiratory tract protection probability. Overall, these preclinical data confirm efficacy of a one-dose vaccine regimen with Ad26.COV2.S in this G614 spike SARS-CoV-2 virus variant Syrian hamster model, show the added benefit of a second vaccine dose, and demonstrate that there are no signs of VAERD under conditions of suboptimal immunity.

Fig. 1. Titration of SARS-CoV-2 challenge dose and characterization of histopathology in Syrian hamsters.

Syrian hamsters (N = 12 per group) were inoculated intranasally with 10², 10^3.3, 10^4.6, or 10^5.9 TCID₅₀ SARS-CoV-2 BetaCoV/Munich/BavPat1/2020 or mock-inoculated with Vero E6 cell-supernatant_. Daily throat swabs were taken, and 2, 3, 4, and 7 days p.i., three hamsters per group were sacrificed and nose and lung tissue collected for virological analysis and histopathology. Replication-competent viral load in a lung tissue, b nose tissue, and c throat swabs was determined by TCID₅₀ assay on Vero E6 cells. LLOD was calculated per animal per gram or milliliter of tissue, and animals with a response at or below the LLOD are shown as open symbols. d Lung tissue was analyzed and scored for presence and severity of alveolitis, alveolar damage, alveolar edema, alveolar hemorrhage, type II pneumocyte hyperplasia, bronchitis, bronchiolitis, peribronchial and perivascular cuffing. Sum of scores are presented as sum of LRT disease parameters (potential range: 0–24). e Nose tissue was analyzed and scored for severity of rhinitis on a scale from 0 to 3. Dotted lines indicate the minimal and maximal scores of histopathology. Median responses per group are indicated with horizontal lines and error bars in panel c indicate the range. p.i. post inoculation, LLOD lower limit of detection, LRT lower respiratory tract, N number of animals, TCID₅₀/g 50% tissue culture infective dose per gram tissue, TCID₅₀/mL 50% tissue culture infective dose per milliliter sample, VP virus particles.

Fig. 2. SARS-CoV-2 neutralizing antibody response elicited by one- and two-dose Ad26.COV2.S vaccine regimes in Syrian hamsters.

a Syrian hamsters were immunized with either 10⁹ or 10¹⁰ VP (N = 12 per dose level) of Ad26-based vaccines, or with 10¹⁰ VP of an Ad26 vector without gene insert as control (Ad26.empty, N = 6). Four weeks after immunization half the hamsters per group received a second immunization with the same Ad26-based vaccine (N = 6 per group). b SARS-CoV-2 neutralization titers were measured 4 weeks after dose 1 and c 4 weeks after dose 2 by wild-type VNA determining the inhibition of the cytopathic effect of SARS-CoV-2 on Vero E6 cells. The sera from Syrian hamsters immunized with Ad26.Empty were pooled into two groups for negative control samples. Median responses per group are indicated with horizontal lines. Dotted lines indicate the LLOD. Animals with a response at or below the LLOD are displayed as open symbols on the LLOD. CPE cytopathic effect, LLOD lower limit of detection, p.i. post inoculation, VNA virus neutralization assay, VP virus particles.

Fig. 3. Protection against SARS-CoV-2 viral replication in Syrian hamsters immunized with Ad26-based vaccines.

Syrian hamsters were intramuscularly immunized with a one-dose regimen and a two-dose regimen of Ad26.S, Ad26.dTM.PP, Ad26.COV2.S, or Ad26.empty (Ad26 vector not encoding any SARS-CoV-2 antigens). Hamsters received an intranasal inoculation with 10² TCID₅₀ SARS-CoV-2 strain BetaCoV/Munich/BavPat1/2020 4 weeks post-dose 1 (week 4) or 4 weeks post-dose 2 (week 8). a, b Right lung tissue and c, d right nasal turbinates were harvested at the end of the 4-day inoculation phase for viral load analysis. Replication-competent virus was measured by TCID₅₀ assay. e, f Throat swab samples were taken daily after inoculation, and viral load area under the curve during the 4-day follow-up was calculated as TCID₅₀/mL × day. The median viral load per group is indicated with a horizontal line. LLOD was calculated per animal and animals with a response at or below the LLOD are shown as open symbols on the LLOD. Comparisons were performed between the Ad26.S, Ad26.dTM.PP, and Ad26.COV2.S groups across dose level, with the Ad26.empty group by Mann–Whitney U test. Statistical differences indicated by asterisks: *p < 0.05; **p < 0.01. LLOD lower limit of detection, TCID₅₀/g 50% tissue culture infective dose per gram tissue, TCID₅₀/mL 50% tissue culture infective dose per mL sample, VP virus particles.

Fig. 4. Dose responsiveness of Ad26.COV2.S on immunogenicity and lung viral load in hamsters.

Syrian hamsters were intramuscularly immunized with 10⁷, 10⁸, 10⁹, or 10¹⁰ VP of Ad26.COV2.S N = 8 per group, or 10¹⁰ VP Ad26.Irr (an Ad26 vector not encoding any SARS-CoV-2 antigens, N = 8). Four weeks after one immunization, SARS-CoV-2 Spike protein-specific antibody-binding titers as measured by ELISA (a) and SARS-CoV-2-neutralizing antibodies as measured by wtVNA (b) were determined. The median antibody responses per group is indicated with a horizontal line. Dotted lines indicate the LLOD. Animals with a response at or below the LLOD were put on LLOD and are shown as open symbols. Hamsters received intranasal inoculation with 10² TCID₅₀ SARS-CoV-2 strain BetaCoV/Munich/BavPat1/2020 4 weeks post immunization (week 4). Right lung tissue was isolated 4 days after inoculation for virological analysis and immunohistochemistry. c Lung viral load was determined by TCID₅₀ assay on Vero E6 cells. The median viral load per group is indicated with a horizontal line. LLOD was calculated per animal, and animals with a response at or below the LLOD are shown as open symbols. d The presence of SARS-CoV-2 NP was determined by immunohistochemical staining. Comparisons were performed between the Ad26.COV2.S dose level groups, with the Ad26.Irr group by Mann–Whitney U test. Statistical differences indicated by asterisks: *p < 0.05; **p < 0.01; ***p < 0.001. Ad26.Irr Ad26 vector not encoding any SARS-CoV-2 antigens, LLOD lower limit of detection, N number of animals, TCID₅₀/g 50% tissue culture infective dose per gram tissue, VP virus particles, NP nucleocapsid protein, wtVNA wild-type virus neutralization assay.

Fig. 5. No signs of VAERD in Ad26 immunized Syrian hamsters inoculated with SARS-CoV-2.

Four days after IN inoculation with 10² TCID₅₀ SARS-CoV-2 (N = 8 per group), a lung tissue was isolated and scored for presence and severity of alveolitis, alveolar damage, alveolar edema, alveolar hemorrhage, type II pneumocyte hyperplasia, bronchitis, bronchiolitis, peribronchial and perivascular cuffing. Sum of scores are presented as the sum of LRT disease parameters. b Four days after inoculation, nose tissue was isolated and scored for severity of inflammation (rhinitis). Horizontal lines denote a pathology score of 0, indicating no histopathology. Symbols in red denote samples from hamsters with breakthrough lung viral load (>10² TCID₅₀/g). Comparisons were performed between the Ad26.COV2.S dose level groups, with the Ad26.Irr group by Mann–Whitney U test. Statistical differences indicated by asterisks: *p < 0.05; **p < 0.01; ***p < 0.001. Ad26.Irr Ad26 vector not encoding any SARS-CoV-2 antigens, LRT lower respiratory tract, N number of animals, VP virus particles.

Fig. 6. Binding and neutralizing antibodies correlate with protection.

Protection was defined as a viral load below 10² TCID₅₀/g in lung tissue, irrespective of vaccine regimen and dose level (see Figs. 3a, b and 4c). Syrian hamsters were immunized once, or twice, with 10⁷, 10⁸, 10⁹, and 10¹⁰ VP Ad26.CoV2.S (N = 56). Hamsters were inoculated with 10² TCID₅₀ SARS-CoV-2, and 4 days later sacrificed for virological analysis of lung tissue. Prior to virus inoculation serum samples were analyzed for a antibody-binding titers and b virus neutralizing antibodies. Median antibody responses per group is indicated with horizontal lines. Dotted lines indicate the LLOD. c Logistic regression models using Firth’s correction were built with protection outcome as the dependent variable, and binding and neutralizing antibody titers from pooled regimens and dose levels of Ad26.COV2.S as independent variable. Dotted lines indicate the 95% confidence interval. LLOD lower limit of detection, N number of animals, TCID₅₀/g 50% tissue culture infective dose per gram tissue, VP virus particles.