Immunogenicity, safety, and efficacy of sequential immunizations with an SIV-based IDLV expressing CH505 Envs

Abstract

A preventative HIV-1 vaccine is an essential intervention needed to halt the HIV-1 pandemic. Neutralizing antibodies protect against HIV-1 infection in animal models, and thus an approach toward a protective HIV-1 vaccine is to induce broadly cross-reactive neutralizing antibodies (bnAbs). One strategy to achieve this goal is to define envelope (Env) evolution that drives bnAb development in infection and to recreate those events by vaccination. In this study, we report the immunogenicity, safety, and efficacy in rhesus macaques of an SIV-based integrase defective lentiviral vector (IDLV) expressing sequential gp140 Env immunogens derived from the CH505 HIV-1-infected individual who made the CH103 and CH235 bnAb lineages. Immunization with IDLV expressing sequential CH505 Envs induced higher magnitude and more durable binding and neutralizing antibody responses compared to protein or DNA +/- protein immunizations using the same sequential envelopes. Compared to monkeys immunized with a vector expressing Envs alone, those immunized with the combination of IDLV expressing Env and CH505 Env protein demonstrated improved durability of antibody responses at six months after the last immunization as well as lower peak viremia and better virus control following autologous SHIV-CH505 challenge. There was no evidence of vector mobilization or recombination in the immunized and challenged monkeys. Although the tested vaccines failed to induce bnAbs and to mediate significant protection following SHIV-challenge, our results show that IDLV proved safe and successful at inducing higher titer and more durable immune responses compared to other vaccine platforms.

Fig. 1. Magnitude and durability of antibody responses induced by sequential IDLV-CH505 Env immunizations.

a Non-human primate immunization regimen with sequential IDLV-CH505 Env +/− protein. The dose of IDLV, protein and adjuvant used for each immunization as well as the challenge and necropsy schedules are indicated. A group of eleven animals were sequentially immunized with IDLV-GFP as the control arm. The HIV-Envs encoded by IDLV are the same for all the vaccine groups (A, B1, and B2) and the VSV-G serotypes used for vector pseudotyping are the same for all the groups. b ELISA binding of plasma antibodies to CH505 T/F Env at the peak and six months post-each immunization with IDLV-CH505 (group B1 animals) and IDLV-CH505 + protein (group B2 animals). Binding titers measured as concentration in µg/mL starting at a 1:3000 plasma dilution. The dotted gray line indicates the improved durability of antibody responses at six months post-each IDLV-CH505 immunization. Asterisks indicate that a statistically significant improvement in the magnitude of Ab responses was detected between weeks 24 and 97 (p = 0.0234). c ELISA binding of plasma antibodies to CH505 T/F Env at the peak and six months post-each immunization with IDLV-CH505 (group A animals, safety arm). d ELISA binding of plasma antibodies to the CH505w.136 SOSIP trimer of plasma samples collected after immunization with IDLV-CH505w.136 SOSIP. Binding titers measured as Log area under the curve (Log AUC) starting at a 1:30 plasma dilution. No envelope binding was detected in plasma samples collected before the first immunization. Asterisks indicate that a statistically significant improvement in the magnitude of Ab responses was detected between weeks 101 and 113 (p = 0.0156).

Fig. 2. Linear epitope specificities of antibody responses induced by sequential IDLV-CH505 Env immunizations.

gp120 (a) and gp41 (b) binding plots for serum samples from macaques immunized with either IDLV-CH505 alone or IDLV-CH505 + protein at 2 weeks after each immunization. Mean binding responses for each of the vaccine groups are shown. Numbers on the x axis are peptide numbers in the array library. The number on the y axis indicates the magnitude of binding calculated as the log2 fold difference, post-/pre-immunization intensity. Different colors of bars represent different strains/clades as indicated in the keys in the panels (TH023, A244, ZM651, TV-1, 1086C, MN, CH505 for panel a and AG, AE, D, M, C, B and A consensus for (b)).

Fig. 3. Linear epitope binding to CH505 sequences.

The heat maps show gp120 and gp41 binding, at week 26 (a) and week 113 (b) post IDLV +/− protein immunization, to the different CH505 strain sequences used in immunization. Binding intensity is shown for each peptide, corrected with its own background value. Light gray shaded areas indicate sequence not present in the array library.

Fig. 4. Magnitude and durability of antibody responses induced by sequential IDLV, protein, and DNA +/− protein immunizations.

a Non-human primate immunization regimens with sequential IDLV +/− protein (NHP122 study), protein alone (NHP79 study) and DNA +/− protein (NHP114 study). b ELISA binding of plasma antibodies to CH505 T/F Env at the peak and either 6 or 3 months (for DNA +/− protein) post four immunizations with each vaccine regimen. Binding titers measured as concentration in µg/mL starting at a 1:3000 plasma dilution. Asterisks indicate that a statistically significant difference was detected (p = 0.004 for IDLV-CH505 + protein vs DNA +/− protein and IDLV-CH505 alone vs DNA alone at peak; p = 0.0162 for IDLV-CH505 alone vs DNA + protein at peak; p = 0.0283 for IDLV-CH505 alone vs CH505 protein alone at 6 months; p = 0.0162 for IDLV + protein vs CH505 protein alone at 6 months; p = 0.004 for IDLV-CH505 +/− protein vs DNA alone at the 3 vs 6 months comparison). c Serum neutralization activity against the clade C tier1 virus CH505.w4.3 post four immunizations with each vaccine regimen.

Fig. 5. SHIV challenge outcomes in IDLV-CH505 +/− protein vaccinated animals.

a Six weeks after completion of the scheduled vaccinations (week 117) the 16 macaques in the B1 and B2 vaccine groups and the 11 macaques in the B3 control group were challenged intrarectally with the Clade C tier 2 SHIV.CH505.375H.dCT. The four animals in group A were challenged once intravenously. The Kaplan–Meier plot shows the percentage of uninfected macaques after 10 weekly IR challenges in each group. b Peak viral load of the infected macaques from the vaccine and control groups. Lines indicate geometric means (p = 0.0506, Exact-Wilcoxon test). c–e Viral load trends from time to infection in each group. Each line represents one animal.

Fig. 6. Env and Gag-specific T-cell responses induced by IDLV-CH505 +/− protein and IDLV-GFP immunizations.

a–c SIV-Gag and d–e CH505 Envs specific T cell responses were measured by IFN-γ ELISPOT in both the IDLV-CH505 +/− protein and IDLV-GFP vaccinated animals. Results are expressed as IFN-γ secreting cells, measured as spot forming cells (SFC) per million PBMC, at different time points post vaccination. Each line represents one animal. f Group means IFN-γ ELISPOT anti-Env responses over time. Error bars indicate S.E.M.

Fig. 7. Magnitude and specificity of antibody responses in IDLV + protein vaccinated animals correlate with time to virus acquisition.

a ELISA binding of plasma antibodies to CH505 T/F Env at the peak and six months post-each immunization with IDLV-CH505 + protein (group B2 animals). Each dot represents one animal. b Number of challenges to virus acquisition for each animal in groups B1 and B2. c Serum neutralization activity against the clade C tier1 virus CH505.w4.3 at the indicated time points. d Comparison of linear epitope specificities between animals in B1 and B2 vaccine groups (red dots), animals with viral loads above or below 250 copies/mL (blue dots) and animals that required either <6 or >6 challenges to become infected (yellow dots). Significant differences fall above the horizontal line and are listed in the text box below the figure for each comparison. e The peptide range and the aa position for each epitope are indicated.

Fig. 8. Absence of vector mobilization or recombination in IDLV-CH505 vaccinated animals challenged intravenously with SHIV-CH505.

a Viral load trends from time of infection in group A animals (IDLV-CH505 alone). Each line represents one animal. b Number of SHIV.CH505.375H.dCT sequences amplified from either PBMC or lymph nodes samples for each animal. c Neighbor-joining phylogenetic tree including CH505 env sequences amplified from either PBMC (full circles) or lymph nodes (full diamonds) from each animal. Each color represents one animal. Bootstrap values ≥80 are shown. Genetic distance is indicated at the bottom of the figure and represents the number of nucleotide substitutions per site.