TL;DR
This study compares the immune responses generated by two different HIV vaccine prime-boost strategies: one using an ALVAC vector and monomeric gp120 protein, and the other using an adenovirus vector and trimeric gp140 protein. While both regimens induced antibody responses, the adenovirus-based approach led to more durable and higher magnitude V1V2 and IgG3 responses, and also elicited CD8+ T-cell responses, which were absent in the ALVAC group. These distinct immune profiles, despite the ultimate lack of efficacy in larger trials, offer crucial insights into what types of immune responses might be necessary for effective HIV prevention.
What Problem Does This Paper Address?
Developing an effective HIV vaccine remains a critical global health challenge. Despite numerous efforts, no vaccine has yet achieved significant protection against HIV acquisition. Two promising vaccine concepts, which were advanced to efficacy trials (HVTN 702 and HVTN 705), ultimately failed to prevent infection. These concepts involved different prime-boost strategies: HVTN 100 used an ALVAC vector prime with a monomeric gp120 protein boost, while HVTN 117/HPX2004 employed an adenovirus vector prime with a trimeric gp140 protein boost. While these efficacy trials provided disappointing results, their earlier phase studies (HVTN 100 and HVTN 117/HPX2004) generated valuable immunological data. This paper aims to dissect and compare the detailed immune responses elicited by these two distinct vaccine regimens from these early phase trials. By understanding the differences and similarities in the induced humoral (antibody) and cellular (T-cell) immunity, researchers hope to gain insights into why these vaccines failed and guide the development of future, more effective HIV vaccine candidates.
Methods
The study employed a cross-protocol analysis comparing immune responses from participants in two Phase 1/2a trials: HVTN 100 and HVTN 117/HPX2004.
- Immunogenicity Assays:
- Binding Antibody Multiplex Assay (BAMA): Used to measure serum HIV-1 total IgG and specific IgG3 binding antibody responses against various Env antigens (gp120, gp140, V1V2, gp70-BCaseA). This assay also allowed for the calculation of an "IgG breadth score" to assess the range of antibodies produced.
- Intracellular Cytokine Staining (ICS): A standard method to measure CD4+ and CD8+ T-cell responses. Specifically, it quantified T-cells expressing key cytokines like IL-2 and IFN-γ in response to vaccine-matched peptide pools.
- Combinatorial Polyfunctionality Analysis of Antigen-Specific T-cell Subsets (COMPASS): A more advanced analysis used to assess the polyfunctionality of T-cell responses. This method goes beyond simply measuring cytokine production to evaluate the combination of cytokines produced by individual T-cells, providing a more nuanced understanding of T-cell activation and function. COMPASS-derived functionality scores (FS and PFS) were calculated.
- Statistical Analysis: The primary analysis utilized super learning and targeted minimum loss estimation (TMLE) to estimate and compare mean response rates and magnitudes, adjusting for potential confounding factors like sex, age, and BMI. Standard statistical tests (Barnard's test, Wilcoxon rank-sum tests) were also used for covariate-unadjusted comparisons and to compare baseline characteristics.
Datasets
The study analyzed data from participants in two clinical trials:
- HVTN 100: A Phase 1/2 study involving participants from South Africa.
- HVTN 117/HPX2004: A Phase 1/2 study involving participants from the United States and Rwanda.
The analysis focused on the per-protocol cohorts of vaccine recipients from these trials. No new datasets were introduced.
Key Insights and Ideas
- Distinct Antibody Profiles: While both regimens induced broad IgG responses to gp120 and gp140, the adenovirus-prime/trimeric gp140-boost (HVTN 117/HPX2004) regimen elicited significantly more durable and higher magnitude IgG responses targeting the V1V2 region of the HIV envelope. This is particularly interesting because V1V2 antibodies have been identified as correlates of risk in previous HIV vaccine trials.
- IgG3 Dominance in Adenovirus Regimen: The adenovirus-based regimen (HVTN 117/HPX2004) also demonstrated a clear advantage in inducing IgG3 responses, especially against consensus gp140 antigens. Higher IgG3 responses are often associated with more potent neutralizing activity.
- CD8+ T-cell Response Gap: A striking difference was the near absence of HIV-specific CD8+ T-cell responses in the ALVAC-prime/monomeric gp120-boost (HVTN 100) group, whereas the adenovirus-based regimen (HVTN 117/HPX2004) successfully elicited these responses. CD8+ T-cells are crucial for clearing infected cells, and their induction is a key goal of many vaccine strategies.
- Polyfunctionality Nuances: While HVTN 100 induced higher polyfunctional CD4+ T-cell responses to Env peptides, HVTN 117/HPX2004 showed higher polyfunctional CD4+ responses to Gag peptides. This highlights how different vaccine platforms can preferentially stimulate responses against different viral components.
- Implications for Future Design: The findings suggest that the specific choice of vaccine platform, vector, and antigen formulation significantly impacts the type and magnitude of immune responses. The more robust and broader responses seen with the adenovirus-based regimen, particularly in V1V2 and CD8+ T-cells, might be more desirable for future vaccine development, even though they did not translate to efficacy in the larger trials.
Key Results
- IgG Breadth: Both regimens induced high IgG breadth against gp120, gp140, and V1V2 antigens by month 12.5/13.
- IgG V1V2 Durability: IgG V1V2 responses were more durable with the HVTN 117/HPX2004 regimen, with significantly higher response rates at month 18 (61.9% vs. 17.8% for gp70-BCaseA V1V2).
- IgG3 Responses: HVTN 117/HPX2004 showed significantly higher and more durable IgG3 responses to consensus Env antigens (e.g., gp140).
- CD4+ T-cell Responses: Polyfunctional CD4+ Env responses were significantly higher in HVTN 100, while polyfunctional CD4+ Gag responses were higher in HVTN 117/HPX2004.
- CD8+ T-cell Responses: CD8+ T-cell responses were almost exclusively observed in HVTN 117/HPX2004, reaching up to 42% response rates, while virtually none were seen in HVTN 100.
Limitations and Future Work
The authors acknowledge several limitations:
- Non-concurrent Assay Runs: Specimens from the two trials were not run on the assays concurrently, which could introduce variability.
- Assay Variations: While qualified assays were used, potential variations over time could impact results.
- Excluded Endpoints: Certain potentially interesting immunogenicity endpoints, such as ADCC responses and avidity, could not be compared due to assay limitations or unavailability. IgA responses were also not compared at the same dilution.
- Vaccination Schedule Differences: HVTN 100 had an additional priming vaccination at month 1, which could contribute to observed differences at month 6.5.
- Specimen Collection Timing: Differences in the timing of specimen collection after vaccination (2 weeks for HVTN 100 vs. 4 weeks for HVTN 117/HPX2004) could influence results.
- Population Differences: The studies were conducted in different geographic regions with distinct populations, and while demographic differences were adjusted for, unmeasured country-specific effects cannot be fully disentangled.
Future Work: The authors suggest that future research should focus on understanding the structural basis of antibody recognition, particularly for the V1V2 region, and explore why the HVTN 100 and HVTN 117/HPX2004 regimens did not elicit more frequent high IgG and IgG3 V1V2 response phenotypes. They also emphasize the need to understand how specific T-cell subsets drive differences in polyfunctionality scores, which could be critical for future vaccine development. The inability of current vaccine platforms to elicit neutralizing antibodies against exposing viral strains remains a significant hurdle.
Summary updated