Vaccine protection against acquisition of neutralization-resistant SIV challenges in rhesus monkeys

Abstract

Preclinical studies of human immunodeficiency virus type 1 (HIV-1) vaccine candidates have typically shown post-infection virological control, but protection against acquisition of infection has previously only been reported against neutralization-sensitive virus challenges. Here we demonstrate vaccine protection against acquisition of fully heterologous, neutralization-resistant simian immunodeficiency virus (SIV) challenges in rhesus monkeys. Adenovirus/poxvirus and adenovirus/adenovirus-vector-based vaccines expressing SIV(SME543) Gag, Pol and Env antigens resulted in an 80% or greater reduction in the per-exposure probability of infection against repetitive, intrarectal SIV(MAC251) challenges in rhesus monkeys. Protection against acquisition of infection showed distinct immunological correlates compared with post-infection virological control and required the inclusion of Env in the vaccine regimen. These data demonstrate the proof-of-concept that optimized HIV-1 vaccine candidates can block acquisition of stringent, heterologous, neutralization-resistant virus challenges in rhesus monkeys.

Figure 1. Immunogenicity and protective efficacy of the adenovirus/poxvirus vaccines

a, Cellular immune responses to SIVsmE543 and SIVmac239 Gag, Pol, and Env as determined by IFN-γ ELISPOT assays at weeks 0, 10, 24, 26, and 52. b, CD8+ and CD4+ total, central/transitional memory (CM; CD28+CD95+), and effector memory (EM; CD28-CD95+) responses to Gag, Pol, and Env as determined by multiparameter IFN-γ ICS assays at week 26. c, SIVmac251 Env ELISAs at weeks 0, 10, 24, 28, and 52. d, SIVsmE660 and SIVmac251 tier 1 pseudovirus NAb assays at weeks 0, 28, and 52. Error bars reflect s.e.m. e, Number of challenges required for acquisition of infection in each vaccine group. f, Statistical analyses include the number of challenges required for 50% infection, hazard ratios with 95% confidence intervals (CI), per-exposure vaccine efficacy (VE), and per-exposure risks of infection in each group. P-values reflect Wald tests using a proportional hazard model. g, Log SIV RNA copies/ml are depicted for each vaccine group at viral setpoint (day 84). ** P=0.0037, Wilcoxon rank-sum tests. The horizontal lines reflect mean setpoint log viral loads.

Figure 1. Immunogenicity and protective efficacy of the adenovirus/poxvirus vaccines

a, Cellular immune responses to SIVsmE543 and SIVmac239 Gag, Pol, and Env as determined by IFN-γ ELISPOT assays at weeks 0, 10, 24, 26, and 52. b, CD8+ and CD4+ total, central/transitional memory (CM; CD28+CD95+), and effector memory (EM; CD28-CD95+) responses to Gag, Pol, and Env as determined by multiparameter IFN-γ ICS assays at week 26. c, SIVmac251 Env ELISAs at weeks 0, 10, 24, 28, and 52. d, SIVsmE660 and SIVmac251 tier 1 pseudovirus NAb assays at weeks 0, 28, and 52. Error bars reflect s.e.m. e, Number of challenges required for acquisition of infection in each vaccine group. f, Statistical analyses include the number of challenges required for 50% infection, hazard ratios with 95% confidence intervals (CI), per-exposure vaccine efficacy (VE), and per-exposure risks of infection in each group. P-values reflect Wald tests using a proportional hazard model. g, Log SIV RNA copies/ml are depicted for each vaccine group at viral setpoint (day 84). ** P=0.0037, Wilcoxon rank-sum tests. The horizontal lines reflect mean setpoint log viral loads.

Figure 2. Correlates of protection against acquisition of infection and virologic control with the adenovirus/poxvirus vaccines

Correlation of (a) log ELISA titers immediately prior to challenge and (b) log tier 1 NAb titers immediately prior to challenge with the number of challenges required to establish infection. Correlation of (c) Gag ELISPOT breadth prior to challenge and (d) Gag ELISPOT magnitude prior to challenge with setpoint viral loads following challenge. Correlates analyses included the 32 vaccinated monkeys (a, b) or the 29 vaccinated animals that became infected (c, d) and did not include the sham controls. P-values reflect Spearman rank-correlation tests. e, V2-specific binding antibodies assessed by surface plasmon resonance response units (RU) for each vaccine group at week 30. * P=0.002, ** P=0.0007, Wilcoxon rank-sum tests. The horizontal lines reflect mean responses. f, Correlation of V2-specific antibody responses with the number of challenges required to establish infection.

Figure 2. Correlates of protection against acquisition of infection and virologic control with the adenovirus/poxvirus vaccines

Correlation of (a) log ELISA titers immediately prior to challenge and (b) log tier 1 NAb titers immediately prior to challenge with the number of challenges required to establish infection. Correlation of (c) Gag ELISPOT breadth prior to challenge and (d) Gag ELISPOT magnitude prior to challenge with setpoint viral loads following challenge. Correlates analyses included the 32 vaccinated monkeys (a, b) or the 29 vaccinated animals that became infected (c, d) and did not include the sham controls. P-values reflect Spearman rank-correlation tests. e, V2-specific binding antibodies assessed by surface plasmon resonance response units (RU) for each vaccine group at week 30. * P=0.002, ** P=0.0007, Wilcoxon rank-sum tests. The horizontal lines reflect mean responses. f, Correlation of V2-specific antibody responses with the number of challenges required to establish infection.

Figure 3. Immunogenicity and protective efficacy of the adenovirus/adenovirus vaccines

a, Cellular immune responses to SIVsmE543 and SIVmac239 Gag, Pol, and Env as determined by IFN-γ ELISPOT assays at weeks 0, 4, 24, 28, and 52. b, CD8+ and CD4+ total, central/transitional memory (CM; CD28+CD95+), and effector memory (EM; CD28-CD95+) responses to Gag, Pol, and Env as determined by multiparameter IFN-γ ICS assays at week 28. c, SIVmac251 Env ELISAs at weeks 0, 4, 24, 28, and 52. d, SIVsmE660 and SIVmac251 tier 1 pseudovirus NAb assays at weeks 0, 28, and 52. Error bars reflect s.e.m. e, Number of challenges required for acquisition of infection in each vaccine group. f, Statistical analyses include the number of challenges required for 50% infection, hazard ratios with 95% confidence intervals (CI), per-exposure vaccine efficacy (VE), and per-exposure risks of infection in each group. P-values reflect Wald tests using a proportional hazard model. g, Log SIV RNA copies/ml are depicted for each vaccine group at viral setpoint (day 84). ** P<0.001, Wilcoxon rank-sum tests. The horizontal lines reflect mean setpoint log viral loads.

Figure 3. Immunogenicity and protective efficacy of the adenovirus/adenovirus vaccines

a, Cellular immune responses to SIVsmE543 and SIVmac239 Gag, Pol, and Env as determined by IFN-γ ELISPOT assays at weeks 0, 4, 24, 28, and 52. b, CD8+ and CD4+ total, central/transitional memory (CM; CD28+CD95+), and effector memory (EM; CD28-CD95+) responses to Gag, Pol, and Env as determined by multiparameter IFN-γ ICS assays at week 28. c, SIVmac251 Env ELISAs at weeks 0, 4, 24, 28, and 52. d, SIVsmE660 and SIVmac251 tier 1 pseudovirus NAb assays at weeks 0, 28, and 52. Error bars reflect s.e.m. e, Number of challenges required for acquisition of infection in each vaccine group. f, Statistical analyses include the number of challenges required for 50% infection, hazard ratios with 95% confidence intervals (CI), per-exposure vaccine efficacy (VE), and per-exposure risks of infection in each group. P-values reflect Wald tests using a proportional hazard model. g, Log SIV RNA copies/ml are depicted for each vaccine group at viral setpoint (day 84). ** P<0.001, Wilcoxon rank-sum tests. The horizontal lines reflect mean setpoint log viral loads.

Figure 4. Correlates of protection against acquisition of infection with the adenovirus/adenovirus vaccines

Correlation of (a) log ELISA titers immediately prior to challenge, (b) log tier 1 NAb titers immediately prior to challenge, (c) V2-specific antibody responses, and (d) rectal IgG antibody responses with the number of challenges required to establish infection. Correlates analyses included the 16 Gag-Pol-Env vaccinated monkeys and did not include the Gag-Pol vaccinated monkeys or the sham controls. P-values reflect Spearman rank-correlation tests.