Effect of HIV Envelope Vaccination on the Subsequent Antibody Response to HIV Infection

Zanele Ditse, Nonhlanhla N. Mkhize, Michael T. Yin, Michael C. Keefer, David C. Montefiori, Georgia D. Tomaras, Gavin Churchyard, Kenneth H. Mayer, Shelly Karuna, Cecilia Morgan, Linda‐Gail Bekker, Koleka Mlisana, Glenda Gray, Zoe Moodie, Peter B. Gilbert, Penny L. Moore, Carolyn Williamson, Lynn MorrismSphere 2020 · DOI

TL;DR

This study investigated whether prior HIV envelope vaccination influences the antibody response after breakthrough HIV infection. Researchers found that existing envelope-based vaccines did not significantly alter the binding or neutralizing antibody responses post-infection, nor did they accelerate the development of broad neutralization. This suggests that current vaccine strategies may need to focus on eliciting different types of immune responses to achieve effective protection.

What Problem Does This Paper Address?

Developing an effective HIV vaccine remains a critical global health challenge. While many studies have characterized the immune responses induced by experimental HIV vaccines in uninfected individuals, there's a significant knowledge gap regarding how these pre-existing vaccine-induced responses behave after an individual becomes infected with HIV. Understanding this is crucial because breakthrough infections, though undesirable, offer a unique opportunity to study how the immune system interacts with both the vaccine and the actual virus. Specifically, researchers wanted to know if prior vaccination could "prime" the immune system in a way that leads to a more robust or broader antibody response upon natural infection, potentially offering some level of control or insight into vaccine design.

Methods

The study employed standard immunological assays to measure antibody responses. Serum samples from participants who acquired HIV infection after being enrolled in HIV vaccine trials were collected longitudinally (annually for up to 6 years post-infection).

  • Binding Antibody Assays: Enzyme-linked immunosorbent assays (ELISA) and a binding antibody multiplex assay (BAMA) were used to detect and quantify antibodies that bind to various HIV envelope proteins and peptides (including gp41, V3, V2, membrane-proximal external region (MPER), and CD4 binding site).
  • Neutralization Assays: The TZM-bl cell neutralization assay was used to assess the ability of antibodies to neutralize HIV. This involved measuring the reduction in luciferase gene expression after infection with pseudotyped viruses. They tested neutralization against "tier 1A" viruses (easier to neutralize) and a panel of "tier 2" viruses (more difficult to neutralize and considered more relevant for broad neutralization).

Datasets

The study utilized serum samples from participants in three specific HIV vaccine trials conducted in South Africa:

  • HVTN 073: Phase 1 trial of a DNA/MVA vaccine (n=1 infected participant).
  • HVTN 086: Phase 1 trial of a DNA/MVA/gp140 protein vaccine (n=2 infected participants).
  • HVTN 204: Phase 2 trial of a DNA/adenovirus serotype 5 (Ad5) vaccine (n=7 infected participants).

Additionally, serum samples from 4 placebo recipients from these trials who also acquired HIV infection were included. Unvaccinated HIV-infected individuals from the CAPRISA cohort served as a control group for comparison of neutralization breadth.

Key Insights and Ideas

  • Waning of Vaccine Immunity: Antibody responses elicited by the vaccines had largely waned by the time of HIV infection, as no significant responses were detected just prior to infection. This highlights the transient nature of some vaccine-induced antibody responses.
  • HIV Infection as the Primary Stimulus: The development of HIV-specific antibody responses post-infection appeared to be primarily driven by the natural infection itself, rather than being significantly boosted by pre-existing vaccine-induced memory. The kinetics and levels of antibody responses were similar between vaccinated and placebo groups.
  • Lack of Priming for Broad Neutralization: The tested envelope-containing vaccines did not prime the immune system to develop tier 2 virus-neutralizing antibodies or accelerate the development of neutralization breadth, which are considered crucial for effective HIV prevention.
  • No Negative Impact: A significant finding is that prior vaccination did not appear to have a negative impact on the immune response to HIV infection. This is an important observation for future vaccine development.

Key Results

  • Binding Antibodies: HIV-specific binding antibodies to gp41, V3, V2, MPER, and the CD4 binding site were detected within the first year of infection. These levels were similar in both vaccinated and placebo recipients and persisted for at least 3 years.
  • Tier 1A Neutralization: Neutralizing antibodies against tier 1A viruses were detected in all participants post-infection, with the highest titers against a subtype C virus (MW965.26). There were no significant differences in these responses between vaccinated and placebo groups.
  • Tier 2 Neutralization: Sporadic and weak neutralization activity against tier 2 viruses was observed in some participants after 2-3 years of infection. However, these responses were similar across vaccinated, placebo, and unvaccinated control groups, indicating no significant priming effect.
  • Neutralization Breadth: Only a small proportion of participants (both vaccinated and unvaccinated) developed neutralization breadth (defined as neutralizing 5 or more viruses). Prior vaccination did not accelerate this development.

Limitations and Future Work

  • Small Sample Size: The study involved a limited number of participants (14 infected vaccine recipients and 4 placebo recipients), which may have restricted the ability to detect subtle differences.
  • Varied Vaccine Regimens: Participants received different vaccine regimens (DNA, MVA, Ad5, protein boosts), which likely resulted in varying levels of antigen exposure and antibody responses, making direct comparisons complex.
  • Time to Infection: A gap of up to one year between estimated infection dates and diagnosis dates could have influenced the measurements.
  • Antigenic Overlap: The study suggests that the B-cell lineages induced by the vaccines might not overlap with those activated by natural HIV infection, which would require further investigation into B-cell receptor repertoires.

The authors suggest that future work should focus on testing newer immunogens that can stimulate B-cell lineages capable of maturing into broadly neutralizing antibodies, potentially through germline-targeting vaccines or trimeric envelope glycoproteins. They also highlight the need for larger efficacy trials with more available samples from vaccinated and infected individuals to expand this type of analysis.

Summary updated

Abstract

Analysis of breakthrough HIV-1 infections could elucidate whether prior vaccination primes relevant immune responses. Here, we measured HIV-specific antibody responses in 14 South African volunteers who acquired HIV infection after participating in phase 1/2 trials of envelope-containing immunogens. Serum samples were collected annually following HIV-1 infection from participants in trials HVTN 073 (subtype C, DNA/MVA, phase 1 trial, <i>n</i> = 1), HVTN 086 (subtype C, DNA/MVA/gp140 protein, phase 1 trial, <i>n</i> = 2), and HVTN 204 (multisubtype, DNA/adenovirus serotype 5 [Ad5], phase 2 trial, <i>n</i> = 7) and 4 placebo recipients. Binding and neutralizing antibody responses to Env proteins and peptides were determined pre- and post-HIV infection using an enzyme-linked immunosorbent assay and the TZM-bl cell neutralization assay, respectively. HIV-infected South African individuals served as unvaccinated controls. Binding antibodies to gp41, V3, V2, the membrane-proximal external region (MPER), and the CD4 binding site were detected from the first year of HIV-1 subtype C infection, and the levels were similar in vaccinated and placebo recipients. Neutralizing antibody responses against tier 1A viruses were detected in all participants, with the highest titers being to a subtype C virus, MW965.26. No responses were observed just prior to infection, indicating that vaccine-primed HIV-specific antibodies had waned. Sporadic neutralization activity against tier 2 isolates was observed after 2 to 3 years of HIV infection, but these responses were similar in the vaccinated and placebo groups as well as the unvaccinated controls. Our data suggest that prior vaccination with these immunogens did not alter the antibody responses to HIV-1 infection, nor did it accelerate the development of HIV neutralization breadth.<b>IMPORTANCE</b> There is a wealth of information on HIV-specific vaccine-induced immune responses among HIV-uninfected participants; however, data on immune responses among participants who acquire HIV after vaccination are limited. Here we show that HIV-specific binding antibody responses in individuals with breakthrough HIV infections were not affected by prior vaccination with HIV envelope-containing immunogens. We also found that these vectored vaccines did not prime tier 2 virus-neutralizing antibody responses, which are thought to be required for prevention against HIV acquisition, or accelerate the development of neutralization breadth. Although this study is limited, such studies can provide insights into whether vaccine-elicited antibody responses are boosted by HIV infection to acquire broader neutralizing activity, which may help to identify antigens relevant to the design of more effective vaccines.
BibTeX
@inproceedings{ditse2020effect,
  author = {Zanele Ditse and Nonhlanhla N. Mkhize and Michael T. Yin and Michael C. Keefer and David C. Montefiori and Georgia D. Tomaras and Gavin Churchyard and Kenneth H. Mayer and Shelly Karuna and Cecilia Morgan and Linda‐Gail Bekker and Koleka Mlisana and Glenda Gray and Zoe Moodie and Peter B. Gilbert and Penny L. Moore and Carolyn Williamson and Lynn Morris},
  title = {{Effect of HIV Envelope Vaccination on the Subsequent Antibody Response to HIV Infection}},
  booktitle = {mSphere},
  year = {2020},
  doi = {10.1128/msphere.00738-19},
}
Figure