Antibody effector analysis of prime versus prime-boost immunizations with a recombinant measles-vectored chikungunya virus vaccine

Abstract

Chikungunya is a mosquito-borne disease that causes periodic but explosive epidemics of acute disease throughout the tropical world. Vaccine development against chikungunya virus (CHIKV) has been hampered by an inability to conduct efficacy trials due to the unpredictability of CHIKV outbreaks. Therefore, immune correlates are being explored to gain inference into vaccine-induced protection. This study is an in-depth serological characterization of Fab- and Fc-mediated antibody responses in selected phase II clinical trial participants following immunization with the recombinant measles-vectored CHIKV vaccine, MV-CHIK. Antibody comparisons were conducted between participants who received prime and those who received prime-boost vaccine regimens. MV-CHIK vaccination elicited potent Fab-mediated antibody responses (such as CHIKV-specific IgG, neutralization, and avidity), including dominant IgG3 responses, which translated into strong antibody-dependent cellular cytotoxicity and antibody-dependent cellular phagocytosis. At 1 month, prime-boost immunization led to significantly greater responses in every measured Fab and Fc antibody parameter. Interestingly, prime-boost-elicited antibodies decreased rapidly over time, until at 6 months both vaccine regimens displayed similar antibody profiles. Nonetheless, antibody avidity and antibody-dependent cellular phagocytosis remained significantly greater following boost immunization. Our observations suggest that a prime-boost administration of MV-CHIK will be more appropriate for CHIKV-endemic regions, while a prime-only regimen may be sufficient for travel purposes or outbreak situations.

Trial registration: ClinicalTrials.gov NCT02861586.

Keywords: Adaptive immunity; Immunology; Vaccines.

Conflict of interest statement

Conflict of interest: This study was partially funded by Themis Bioscience GmbH, a subsidiary of Merck & Co. Inc. RT and KR are employees of Themis Bioscience GmbH.

Figures

Figure 1. Study design and characterization of the CHIK-specific IgG response elicited by MV-CHIK prime and prime-boost vaccination strategies.

(A) Participants were vaccinated with either MV-CHIK prime at day 0 (P only) or prime (at month –1) and boost (at day 0) (P+B), and serum antibody responses were compared at 1 and 6 months after vaccination. MV-vaccinated groups served as controls. (B–F) Total IgG binding to CHIKV virus was measured using ELISA and compared (B) between groups and (C) at 1 and 6 months for paired MV-CHIK–vaccinated samples. (D) Fold change in IgG (mean ± 95% CI) in response to CHIKV virus was compared between MV-CHIK P only and P+B. IgG binding to 2 CHIK surface glycoproteins, E1 and E2, was measured using Luminex immunoassay and compared between groups at (E) 1 month and (F) 6 months after vaccination. (G) Distribution of the CHIK antigens targeted by the antibody response was estimated by the ratio of E2-specific to E1-specific IgG (mean ± 95% CI) and compared between prime and prime-boost MV-CHIK vaccinations. IgG responses between groups were compared using Mann-Whitney U test, while paired analysis between groups at 1 and 6 months was conducted using paired Wilcoxon signed-rank test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 2. Avidity and neutralizing capacity of CHIK virus–specific antibody responses elicited by MV-CHIK prime and prime-boost immunization.

(A) Binding strength of CHIK virus–specific antibodies, as indicated by avidity index estimations following 4, 6, and 8 M urea washes. (B) CHIK virus neutralization (measured by PRNT assay) and (C) neutralization normalized to total IgG (mean ± 95% CI) compared between vaccinated groups. (D and E) Longitudinal assessment of the CHIKV-neutralizing antibody response using (D) paired analysis between 1 and 6 months and (E) fold change in CHIKV neutralization from 1 to 6 months (mean ± 95% CI) in MV-CHIK prime (P) and prime-boost (P+B) immunization. Avidity index and PRNT50 titers were compared between groups using Mann-Whitney U test, and analysis between groups at 1 and 6 months was conducted using paired Wilcoxon signed-rank test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 3. IgG subclass distribution of CHIK virus–specific antibody responses following MV-CHIK prime and prime-boost vaccinations.

(A) IgG1, IgG2, IgG3, and IgG4 CHIK virus–binding endpoint titers (measured by ELISA) in MV-CHIK–vaccinated individuals. (B and C) Longitudinal analysis of CHIK virus–specific IgG1 and IgG3. (B) Endpoint titers at 1 and 6 months. (C) Fold change in antibody titers from 1 to 6 months (mean ± 95% CI). (D–G) The ratio of each IgG subclass (IgG1, IgG2, IgG3, and IgG4) to total IgG, and (H) the ratio of IgG1 to IgG2 (mean ± 95% CI). IgG subclass titers were compared between groups using Mann-Whitney U test, and comparisons between groups at 1 and 6 months used paired Wilcoxon signed-rank test (**P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 4. Antibody-dependent cellular phagocytosis of CHIK virus by a monocytic cell line following vaccination with MV-CHIK.

(A) Phagocytic scores (measured using a whole-virus phagocytic assay) compared between vaccinated groups. (B) Paired longitudinal antibody-dependent cellular phagocytosis (ADCP) responses compared at 1 and 6 months. (C) Normalized phagocytic scores (i.e., phagocytic scores normalized to CHIKV-specific IgG titers) compared between P- and P+B MC-CHIK–vaccinated groups. Phagocytic scores and normalized phagocytic scores were compared between groups using Mann-Whitney U test, and paired analysis between groups at 1 and 6 months was conducted using paired Wilcoxon signed-rank test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 5. NK-mediated ADCC following MV-CHIK vaccination.

ADCC was assessed through the proxy measures of ADNKA degranulation (CD107α) and cytokine release (IFN-γ). (A) Percentage of CD107α+ cells and (B) IFN-γ+ cells in vaccinated groups. Paired longitudinal analysis of the percentage of (C) CD107α+ cells and (D) IFN-γ+ cells at 1 and 6 months after vaccination. Normalized percentage of (E) CD107α+ cells and (F) IFN-γ+ cells (to CHIKV-specific IgG titers) in P- and P+B MV-CHIK–vaccinated groups. Percentages and normalized percentages of CD17α+ and IFN-γ+ were compared between groups using Mann-Whitney U test, and paired analysis between groups at 1 and 6 months was conducted using paired Wilcoxon signed-rank test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 6. Multidimensional unsupervised analysis of MV-CHIK–vaccinated groups and measured antibody phenotypes.

(A and B) Principal component analysis (PCA) was conducted on MV-CHIK vaccine–induced antibody responses following prime (P) and prime-boost (P+B) regiments at 1 and 6 months. (A) Individual plot with vaccine groups highlighted in different colors. Ellipses in the PCA individual plot represent the confidence ellipses around the centroid (marked by the larger symbol for each group) of each vaccine group. (B) Circle plot highlighting the variables contributing to the spread of points. (C) Correlation matrix with hierarchical clustering of variables utilized in the PCA analysis. Variables included in the PCA and hierarchical correlation matrix include neutralization (PRNT50 titers); virus-binding total IgG, IgG1, IgG2, IgG3, and IgG4 (endpoint titers); E1-binding IgG (MFI); E2-binding IgG (MFI); avidity at 8 M (AI); ADCP (phagocytic score); and ADCC (ADNKA degranulation as percentage of CD107α+ cells, and ADNKA cytokine release as percentage of IFN-γ+ cells).