Coadministration of CH31 Broadly Neutralizing Antibody Does Not Affect Development of Vaccine-Induced Anti-HIV-1 Envelope Antibody Responses in Infant Rhesus Macaques

Abstract

Prevention of mother-to-child transmission (MTCT) is an indispensable component in combatting the global AIDS epidemic. A combination of passive broadly neutralizing antibody (bnAb) infusion and active vaccination promises to provide protection of infants against MTCT from birth through the breastfeeding period and could prime the immune system for lifelong immunity. In this study, we investigate the impact of a single infusion of CD4 binding site (CD4bs) bnAb administered at birth on de novo antibody responses elicited by concurrent active HIV envelope vaccination. Four groups of infant macaques received active immunizations with subunit Env protein or modified vaccinia Ankara (MVA)-vectored Env and subunit Env protein, with or without a single intravenous coadministration of CH31 bnAb at birth. Vaccinated animals were monitored to evaluate binding and functional antibody responses elicited by the active vaccinations. Despite achieving plasma concentrations that were able to neutralize tier 2 viruses, coadministration of CH31 did not have a large impact on the kinetics, magnitude, specificity, or avidity of vaccine-elicited binding or functional antibody responses, including epitope specificity, the development of CD4bs antibodies, neutralization, binding to infected cells, or antibody-dependent cell-mediated cytotoxicity (ADCC). We conclude that infusion of CD4bs bnAb CH31 at birth does not interfere with de novo antibody responses to active vaccination and that a combination of passive bnAb infusion and active HIV-1 Env vaccination is a viable strategy for immediate and prolonged protection against MTCT.IMPORTANCE Our study is the first to evaluate the impact of passive infusion of a broadly neutralizing antibody in newborns on the de novo development of antibody responses following active vaccinations in infancy. We demonstrated the safety and the feasibility of bnAb administration to achieve biologically relevant levels of the antibody and showed that the passive infusion did not impair the de novo antibody production following HIV-1 Env vaccination. Our study paves the way for further investigations of the combination strategy using passive plus active immunization to provide protection of infants born to HIV-1-positive mothers over the entire period of risk for mother-to-child transmission.

Keywords: AIDS vaccine; HIV; antibody; broadly neutralizing antibody; infant; nonhuman primate; passive immunization; pediatric vaccine.

FIG 1

Animal study design schema. Vaccine components are represented by different symbols. Group identifiers and numbers of animals per groups are indicated at the top of the time scale for each vaccine group.

FIG 2

Level of in vivo CH31 and de novo binding responses in vaccinated macaques. (A) Levels of CH31 MAb in groups that received CH31 coadministration, measured as Env-specific human (hu) IgG. D, day; wk, week. (B) Levels of IgG binding for vaccine strain 1086.C gp120 over time as measured in an anti-monkey IgG ELISA. (C) Magnitude of binding to a panel of HIV-1 Env antigens as measured in BAMA. (D) Magnitude of binding to a panel of region/epitope-specific antigens as measured in BAMA. *, unadjusted P < 0.05; **, unadjusted P < 0.01. All P values are >0.05 once values were adjusted for multiple comparison (Tables 1 S1, and S2 give both unadjusted P and FDR P values for all comparisons). Black horizontal bars represent group median values. For panel D, binding to the V3 Ag panel ($) (calculated as mean MFI for gp70 Con V3, gp70-MN V3, and linear V3.C by each animal) was lower for the Env-only group than for the MVA/Env group at week 2 (unadjusted P < 0.05).

FIG 3

Magnitude of binding to linear epitopes measured in peptide microarray. (A) Heat map showing binding magnitude to 1086.C linear epitopes at week 2 and week 8 by all animals. nd, not done (2 samples were not available for week 2 and, thus, not analyzed in linear epitope mapping). (B) Scatter plot for 1086.C linear epitope binding magnitude for the vaccine groups at week (wk) 2 and week 8. Black cross bars represent group median values. *, unadjusted P < 0.05. All P values were >0.05 once values were adjusted for multiple comparison (Tables 1, S1, and S2 give both unadjusted P and FDR P values for all comparisons).

FIG 4

Functional antibody responses in vaccinated macaques. (A to C) Levels of neutralization (of C.MV965 on TZM-bl cells), infected cell binding (for 1086.C IMC-infected cells), and ADCC (against 1086.C gp120-coated cells) responses over time. (D) Neutralizing activity against MW965 and SO032 at week 14. (E) Levels of infected cell binding (secondary MFI) to a panel of HIV-1 IMC-infected cells at key time points. (F) Levels of infected cell binding (positive percent cells) to a panel of HIV-1 IMC-infected cells at key time points. (G) Levels of ADCC response against 1086.C gp120-coated cells (maximum percent granzyme B activity) over time. *, unadjusted P < 0.05. All P values are >0.05 once values were adjusted for multiple comparison (Tables 1, S1, and S2 give both unadjusted P and FDR P values for all comparisons). For panels A, B, and C, the thin lines represent individual animals while thick lines indicate group median values. For panels D to G, black horizontal bars represent group median values.

FIG 5

Avidity of plasma IgG binding to a panel of Env, V1V2, and V3 antigens at week 8 and week 14 shown as disassociation (k_d) rate (A) and avidity score (B), as measured by SPR. *, unadjusted P < 0.05. All unadjusted P and FDR P values are >0.05 (Tables 1, S1, and S2 give details). Black horizontal bars represent group median values. Avidity data are not available for all samples as avidity could not be measured for samples that did not bind sufficiently to the analytes.

FIG 6

Level of soluble CD4 blocking activity measured in sCD4 blocking ELISA as an indicator for a potential CD4bs binding antibody response. *, unadjusted P < 0.05. All P values are >0.05 once values were adjusted for multiple comparison (Tables 1, S1, and S2 give both unadjusted P and FDR P values for all comparisons). Horizontal black bars represent group median values.