A SARS-CoV-2 Spike Ferritin Nanoparticle Vaccine Is Protective and Promotes a Strong Immunological Response in the Cynomolgus Macaque Coronavirus Disease 2019 (COVID-19) Model

Abstract

The COVID-19 pandemic has had a staggering impact on social, economic, and public health systems worldwide. Vaccine development and mobilization against SARS-CoV-2 (the etiologic agent of COVID-19) has been rapid. However, novel strategies are still necessary to slow the pandemic, and this includes new approaches to vaccine development and/or delivery that will improve vaccination compliance and demonstrate efficacy against emerging variants. Here, we report on the immunogenicity and efficacy of a SARS-CoV-2 vaccine comprising stabilized, pre-fusion spike protein trimers displayed on a ferritin nanoparticle (SpFN) adjuvanted with either conventional aluminum hydroxide or the Army Liposomal Formulation QS-21 (ALFQ) in a cynomolgus macaque COVID-19 model. Vaccination resulted in robust cell-mediated and humoral responses and a significant reduction in lung lesions following SARS-CoV-2 infection. The strength of the immune response suggests that dose sparing through reduced or single dosing in primates may be possible with this vaccine. Overall, the data support further evaluation of SpFN as a SARS-CoV-2 protein-based vaccine candidate with attention to fractional dosing and schedule optimization.

Keywords: Army Liposomal Formulation QS-21; COVID-19; SARS-CoV-2; SpFN; aluminum hydroxide; ferritin nanoparticle; primate; vaccine.

Figure 1

Study design. Vaccinations occurred on Study Days −56 and −28, as indicated by the syringes in the diagram. Cynomolgus macaques were challenged with SARS-CoV-2 (WA-1) by the intranasal (IN) and intratracheal (IT) routes on Study Day 1, as indicated. Days of physical examination and blood/specimen collection are indicated by red arrows, and euthanasia days are indicated by black arrows.

Figure 2

Disease characteristics following SARS-CoV-2 challenge of SpFN-vaccinated cynomolgus macaques (CM). (A) DSI M00 telemetry devices were used to collect body temperature data. The panel on the left shows the maximum change (Δ) in temperature (°C) for the 24 h daily time period by study group. The panel on the right shows the fever-h measurements by study group, which are the sum of the significant temperature elevation values, and give an indication of the intensity of the fever by crudely calculating the area under the curve. Group 1 (red symbols and lines) = SpFN + ALFQ; Group 2 (green symbols and lines) = SpFN + AlOH₃; Group 3 (black symbols and lines) = controls. Error bars represent the standard deviation for a study group on a particular study day. Significant differences (p < 0.05) between Groups 1 and 2, as determined by Wilcoxon rank-sum test, are indicated with a star (*). (B) Radiographic findings are shown. Left panel: CM #7 (Group 1, SpFN + ALFQ), Study Day 3, potential very mild opacity increase in the right middle lung lobe. Middle panel: CM #3 (Group 2, SpFN + AlOH₃), Study Day 5, moderate infiltrates present in the right middle and caudal lobes, with partial obfuscation of the cardiac silhouette on the right. Right panel: CM #17 (Group 3, control), Study Day 5, partial obfuscation of the right cardiac silhouette with worsening of infiltrate and opacity of middle and caudal lobes bilaterally (R > L); infiltrate also present on the left side. (C) Histopathology findings in lung tissue are shown. Left panel: Left cranial lung lobe, CM #9 (Group 1, SpFN + ALFQ), 10× magnification of a peripheral section demonstrating a small amount of perivascular inflammation (black arrow). Middle panel: Left caudal lung lobe, CM #16 (Group 2, SpFN + AlOH₃), 10× magnification of a peripheral section demonstrating more perivascular inflammation. Right panel: Right caudal lung lobe, CM #11 (Group 3, control), 10× magnification of a central section demonstrating significantly more inflammation that expands into alveolar spaces, septa, and perivascular areas than in the previous images from animals in Groups 1 and 2.

Figure 3

Clinical pathology and respiratory tract-associated viral RNA levels following SARS-CoV-2 challenge of SpFN-vaccinated cynomolgus macaques (CM). Data are presented by group. Group 1 (red symbols and lines) = SpFN + ALFQ; Group 2 (green symbols and lines) = SpFN + AlOH₃; Group 3 (black symbols and lines) = controls. Significant differences (p < 0.05) between control and vaccinated, as determined by Wilcoxon rank-sum test, are indicated with a star (*). Horizontal bars represent the mean value. (A) Clinical pathology. Measurements are shown as percent change from baseline (average of values from Study Days −4 and 1 for each animal) for peak values for each analyte. The panel on the left shows hematology data, and the panel on the right shows clinical chemistry data. (B) Subgenomic RNA in nasopharyngeal (NP) swabs (left panel) and BAL (right panel). Data are shown as Log₁₀ copies/mL. The dotted lines demarcate assay lower limit of linear performance range (Log₁₀ = 2.65, corresponding to 450 copies/mL). (C) SARS-CoV-2 RT-qPCR to detect genomic RNA was performed on NP swabs (left panel) and BAL (right panel). Data are shown as Log₁₀ genomic equivalents (ge)/mL. The dotted lines demarcate assay lower limit of linear performance range for BAL (Log₁₀ = 2.65) or the lower limit of quantification for NP swabs (Log₁₀ = 4.00).

Figure 4

Characterization of the total S1-specific antibody response using ELISA, and antigen-specific antibody response by MagPix. Data are presented by group. Group 1 (red symbols and lines) = SpFN + ALFQ; Group 2 (green symbols and lines) = SpFN + AlOH₃; Group 3 (black symbols and lines) = controls. Significant differences (p < 0.05) between control and vaccinated, as determined by Wilcoxon rank-sum test, are indicated with a star (*). (A) Serum samples were assessed using the Euroimmun SARS-CoV-2 S1 IgG or IgA ELISA kit. The dotted line represents the assay cutoff, above which a sample is considered above background noise (i.e., positive). The total S1 IgG response is shown in the left panel, and the total S1 IgA response is shown in the right panel. Black arrows indicate the days of vaccination. OD = optical density (Adjusted OD is the OD resulting following removal of background noise). (B) The IgG (left panels) and IgM (right panels) responses to the indicated antigens were measured by a Magpix immunoassay.

Figure 5

Total serum RBD binding response following SpFN vaccination in cynomolgus macaques (CM). Serum collected from all animals on Study Day −56 or −14 was analyzed by BLI for binding to SARS-CoV-2 WA-1 and VoC RBD molecules. Data are presented by group. Group 1 (Circles) = SpFN + ALFQ; Group 2 (Squares) = SpFN + AlOH₃. The horizontal bars represent the mean for each group. Groups were compared using unpaired non-parametric two-tailed Mann–Whitney test (** p < 0.01, and *** p < 0.001).

Figure 6

Adjuvanted SpFN vaccine-elicited spike and RBD binding and ACE2 inhibitory responses to SARS-CoV-2 assessed by MSD immunoassay. Humoral responses were measured by MSD immunoassay. Data are presented by group. Group 1 (red symbols and lines) = SpFN + ALFQ; Group 2 (green symbols and lines) = SpFN + AlOH₃; Group 3 (black symbols and lines) = controls. Dotted lines represent geometric mean of pre-immune values plus five standard deviations, considered the threshold for positive responses. (A) Serum SARS-CoV-2 (WA-1) S-specific IgG responses are depicted as binding antibody units/mL (BAU/mL), using the WHO International Standard and a conversion factor of 0.009. Thick lines indicate geometric means for each group and thin lines represent individual animals. (B) Serum IgG binding antibody responses to S antigens from SARS-CoV-2 WA-1, Alpha, Beta, and Gamma on Study Day 1. Significant differences between ALFQ and AlOH₃-adjuvanted SpFN are indicated (* p < 0.05, ** p < 0.01, *** p < 0.001; Mann–Whitney test). (C) WA-1 RBD-specific serum IgG binding responses are depicted longitudinally as BAU/mL (conversion factor 0.027). (D) Serum IgG binding antibody responses against RBD antigens from WA-1, Alpha, Beta, and Gamma. Significant differences between ALFQ and AlOH₃-adjuvanted SpFN are indicated. (E) Serum inhibition of SARS-CoV-2 (WA-1) S binding to angiotensin-converting enzyme 2 (ACE2) reported as arbitrary units (AU/mL). (F). Serum inhibition of ACE2 binding to SARS-CoV-2 S proteins from WA-1, Alpha, Beta, and Gamma on Study Day 1. Significant differences between ALFQ and AlOH₃-adjuvanted SpFN are indicated.

Figure 7

SARS-CoV-2 neutralizing antibody responses following SpFN vaccination in cynomolgus macaques (CM). (A) The PRNT₈₀ titers to the WA-1 strain of SARS-CoV-2 are shown. Data are presented by group. Group 1 (red symbols and lines) = SpFN + ALFQ; Group 2 (green symbols and lines) = SpFN + AlOH₃; Group 3 (black symbols and lines) = controls. Significant differences (p < 0.05) between Groups 1 and 2, as determined by Wilcoxon rank-sum test, are indicated with a star (*) (note: Comparison of Group 3 to Groups 1 and 2 revealed statistical significance by Wilcoxon rank-sum test at all post-vaccination time points). Black arrows indicate the days of vaccination. GMT = geometric mean neutralization titer. Error bars represent the geometric standard deviation. (B) The 50% infectious dose (ID50) geometric mean (pseudovirion assay, Alpha and Omicron variants and SARS-CoV-1) and PRNT₈₀ GMTs (live virus assay, Beta, Gamma, and Delta variants) against variants of concern (VoC) are shown. Error bars represent the standard deviation. Statistical relevance was determined using unpaired t tests (* p < 0.05, ** p < 0.01, and *** p < 0.001). In the top panels, Group 1 (SpFN + ALFQ) and Group 2 (SpFN + AlOH₃) titers are compared for each VoC. In the middle panels, titers for Group 1 against WA-1 are compared to those measured for VoC. In the bottom panels, titers for Group 2 against WA-1 are compared to those measured for VoC.

Figure 8

Non-neutralizing antibody effector functions following SpFN vaccination in cynomolgus macaques (CM). Data are presented by group. Group 1 (red symbols and lines) = SpFN + ALFQ; Group 2 (green symbols and lines) = SpFN + AlOH₃; Group 3 (black symbols and lines) = controls. Significant differences (p < 0.05) between control and vaccinated, as determined by Wilcoxon rank-sum test, are indicated with a star (*). (A) ADCP. (B) ADCD. MFI = mean fluorescence intensity.

Figure 9

SARS-CoV-2 and SARS-CoV-1 S-specific CD4+ T cell responses elicited by adjuvanted SpFN vaccination. (A–D). T cell responses were assessed by SARS-CoV-2 spike peptide pool stimulation and intracellular cytokine staining of PBMC collected at Study Day −14 (2 weeks post-boost). S-specific memory CD4+ T cells were defined by expression of: A. Th1 cytokines (IFN-γ, TNF-α and IL-2), B. Th2 cytokines (IL-4 and IL-13), C. IL-21, or D. CD40L. Boolean combinations of cytokine-positive memory T cells were summed for Th1 and Th2 response magnitude. (E) Pie charts depict polyfunctionality of S-specific Th1 CD4+ T cell responses 2 weeks post-boost for the two adjuvanted SpFN vaccine arms, as assessed by Boolean combination gating of IFN-γ, TNF-α, and IL-2 positive cells. COMPASS analysis of S-specific CD4+ T cell functionality (F) and polyfunctionality (G) determined by Th1 cytokine (TNFα, IL2, IFNγ), CD40L, and IL-21 expression. H-I. SARS-CoV-1 S-specific CD4+ T cell responses were measured by PBMC stimulation with SARS-CoV-1 spike peptide pools. The frequency of memory CD4+ T cells expressing Th1 cytokines (IFN-γ, TNF-α, and IL-2) (H) and CD40L (I) is shown. Study groups are depicted as follows: SpFN + ALFQ, red; SpFN + AlOH₃, green; controls, black. Box plot horizontal lines indicate the mean; top and bottom reflect the minimum and maximum. The fraction of animals within each group with a positive response following each vaccination is indicated. Significance was assessed using a Kruskal–Wallis test followed by a Dunn’s post-test to compare all study groups, or a Mann–Whitney test to compare the vaccine groups.