The phase 1/2a APPROACH trial and the parallel NHP 13-19 rhesus monkey study (Barouch et al., 2018) established that a mosaic adenovirus-26 (Ad26)-based HIV-1 vaccine regimen with clade C gp140 protein boosts was immunogenic in humans and reduced the risk of acquisition per exposure by 67% against repeated SHIV-SF162P3 challenges in monkeys, yet induced no broadly neutralizing antibodies against tier-2 viruses. The paper left open three linked questions: whether the monkey challenge model predicts human efficacy, what immune mechanisms actually protect humans, and whether vaccination can elicit broader neutralization or protection must rely on non-neutralizing antibody functions and cellular immunity. The subsequent human efficacy trials of this regimen are now complete and negative: Imbokodo/HVTN 705 (vaccine efficacy 14.1%, 95% CI -22.0 to 39.5) and Mosaico/HVTN 706 (vaccine efficacy -0.7%, 95% CI -50.9 to 32.8, discontinued for insufficient efficacy), empirically answering the human-efficacy branch in the negative and sharply exposing the monkey-human disconnect. The surviving open core — identification and causal validation of mechanistic immune correlates of protection against HIV-1 acquisition in humans, development of challenge models with human predictive validity, and induction of tier-2 broadly neutralizing antibody responses by vaccination (with germline-targeting priming immunogens now reaching early precursor maturation but not breadth) — remains the central unsolved problem of HIV-1 vaccinology.

Background

An effective prophylactic HIV-1 vaccine does not exist. Barouch et al. (2018) reported two parallel studies of 'mosaic' vaccine immunogens — antigens computationally designed from many HIV-1 sequences to cover global viral diversity — delivered by a replication-incompetent adenovirus serotype 26 (Ad26) vector, boosted with injections of clade C gp140, the outer envelope glycoprotein of HIV-1 used as a soluble recombinant protein with aluminium adjuvant. In the phase 1/2a APPROACH trial (393 participants), the Ad26/Ad26 plus high-dose gp140 regimen was the most immunogenic: it elicited envelope-binding antibodies, antibody-dependent cellular phagocytosis (an 'Fc effector' function, meaning antiviral activity exercised by antibodies recruiting other immune cells rather than by directly blocking viral entry), and T-cell responses in most recipients. In the parallel NHP 13-19 study, 72 rhesus monkeys received analogous regimens and were then exposed repeatedly to SHIV-SF162P3, a chimeric simian-human immunodeficiency virus carrying the HIV-1 envelope of the challenge stock; the leading regimen reduced the risk of acquisition per exposure by 67% (p=0.007). Monkey-level correlates of that protection were statistical only: envelope-specific antibody measured by ELISA (enzyme-linked immunosorbent assay) and envelope-specific T cells measured by ELISPOT (a single-cell cytokine secretion assay). The regimen induced no broadly neutralizing antibodies (bnAbs) against 'tier-2' viruses — the standard laboratory classification for primary patient-derived HIV-1 isolates whose conformationally closed envelope spikes are hard to neutralize, in contrast to easily neutralized 'tier-1' lab strains. Because non-neutralizing mechanisms (Fc-mediated effector functions and cytotoxic T cells) were the plausible basis of monkey protection, the authors flagged three uncertainties that structured later work: the predictive validity of the SHIV-SF162P3 rhesus model for human efficacy; the absence of any mechanistic (as opposed to statistical) correlate of protection in humans; and whether vaccine modification can achieve broader neutralization. Context for what followed: the only HIV vaccine trial to date with statistically significant efficacy was RV144 (about 31% in Thailand, 2009), whose signals were also non-neutralizing (IgG antibodies to the V1V2 variable loops of envelope); passive-infusion trials of the bnAb VRC01 (the Antibody-Mediated Prevention trials) later showed only isolate-specific, not overall, efficacy, as reviewed by Kelkar et al. (2025). Since 2023, a 'germline-targeting' strategy has entered human testing: vaccine nanoparticles (such as eOD-GT8 60mer, and mRNA-lipid nanoparticle versions) engineered to bind the unmutated precursors of known bnAb lineages such as the VRC01 class, aiming to eventually shepherd those rare B cells through maturation toward breadth.

Problem Statement

The mosaic Ad26-based HIV-1 vaccine concept has now been tested in human efficacy trials and failed to protect: Imbokodo/HVTN 705 showed 14.1% (95% CI -22.0 to 39.5) vaccine efficacy and Mosaico/HVTN 706 showed -0.7% (95% CI -50.9 to 32.8) vaccine efficacy, even though the same concept reduced per-exposure acquisition risk by 67% against repeated SHIV-SF162P3 challenges in rhesus monkeys, and no vaccine regimen has yet induced broadly neutralizing antibodies against tier-2 HIV-1 isolates. Three linked questions that remain open define the problem: (1) Mechanistic correlates of protection in humans — determine which immune mechanisms (virus-specific T-cell responses, Fc-mediated antibody effector functions such as antibody-dependent cellular cytotoxicity and phagocytosis, V1V2-directed binding antibodies, or neutralizing antibodies) actually prevent HIV-1 acquisition in humans, and validate them causally rather than as statistical correlates of risk; (2) Model validity — establish nonhuman primate challenge models (or other preclinical systems) whose protection outcomes quantitatively predict human vaccine efficacy, so that the monkey-human disconnect demonstrated by this lineage cannot recur; (3) Neutralization breadth — determine whether vaccination can elicit broadly neutralizing antibody responses against tier-2 viruses in humans (with germline-targeting and other immunogen-design strategies currently at the precursor-priming stage), or alternatively demonstrate which non-neutralizing combination of humoral and cellular mechanisms could be engineered to mediate reliable protection. These three questions share one overarching objective — a mechanistically validated design strategy for a protective HIV-1 vaccine, which currently does not exist. They are, however, independent and separately attackable sub-goals rather than interchangeable alternatives: each is posed on its own terms, progress on one does not by itself settle the others, any one of them resolved with a mechanistically justified design implication constitutes a resolution of the problem, and the overarching objective is their unifying rationale rather than a fourth, separately answerable deliverable.

References

  1. Barouch DH, Tomaka FL, Wegmann F, Stieh DJ, Alter G, Robb ML, et al. Evaluation of a mosaic HIV-1 vaccine in a multicentre, randomised, double-blind, placebo-controlled, phase 1/2a clinical trial (APPROACH) and in rhesus monkeys (NHP 13-19). The Lancet 2018;392:232-243. doi:10.1016/S0140-6736(18)31364-3. https://doi.org/10.1016/S0140-6736(18)31364-3
  2. Gray GE, Mngadi K, Lavreys L, Nijs S, Gilbert PB, Hural J, et al. Mosaic HIV-1 vaccine regimen in southern African women (Imbokodo/HVTN 705/HPX2008): a randomised, double-blind, placebo-controlled, phase 2b trial. The Lancet Infectious Diseases 2024;24:1201-1212. doi:10.1016/S1473-3099(24)00358-X. https://doi.org/10.1016/S1473-3099(24)00358-X
  3. Kenny A, van Duijn J, Dintwe O, Heptinstall J, Burnham R, Sawant S, et al. Immune correlates analysis of the Imbokodo (HVTN 705/HPX2008) efficacy trial of a mosaic HIV-1 vaccine regimen evaluated in Southern African people assigned female sex at birth: a two-phase case-control study. eBioMedicine 2024;108:105320. doi:10.1016/j.ebiom.2024.105320. https://doi.org/10.1016/j.ebiom.2024.105320
  4. Voillet V, van Duijn J, Vanshylla K, Dintwe OB, Pattacini L, Omar FL, et al. Unbiased cell clustering analysis of vaccine-induced T cell responses in the Imbokodo HIV-1 vaccine trial. eBioMedicine 2025;122:106017. doi:10.1016/j.ebiom.2025.106017. https://doi.org/10.1016/j.ebiom.2025.106017
  5. Buchbinder SP, Spinosa Guzman S, Sanchez J, Willems W, Stieh DJ, van Duijn J, et al. Efficacy and safety of a mosaic HIV-1 vaccine regimen in men who have sex with men and transgender individuals (HVTN 706/HPX3002/Mosaico): a global, randomised, double-blind, placebo-controlled, phase 3 trial. The Lancet HIV 2025;12:e823-e835. doi:10.1016/S2352-3018(25)00195-X. https://doi.org/10.1016/S2352-3018(25)00195-X
  6. Kelkar NS, Carpenter MC, Ackerman ME. Losing Less in Translation: Relating the Preclinical Evidence Base for Active and Passive Immunity to Prevention of HIV-1 Infection in Humans. Journal of Medical Primatology 2025;54:e70040. doi:10.1111/jmp.70040. https://doi.org/10.1111/jmp.70040
  7. Cohen KW, De Rosa SC, Fulp WJ, deCamp AC, Fiore-Gartland A, Mahoney CR, et al. A first-in-human germline-targeting HIV nanoparticle vaccine induced broad and publicly targeted helper T cell responses. Science Translational Medicine 2023;15:eadf3309. doi:10.1126/scitranslmed.adf3309. https://doi.org/10.1126/scitranslmed.adf3309
  8. Willis JR, Prabhakaran M, Muthui M, Naidoo A, Sincomb T, Wu W, et al. Vaccination with mRNA-encoded nanoparticles drives early maturation of HIV bnAb precursors in humans. Science 2025;389:eadr8382. doi:10.1126/science.adr8382. https://doi.org/10.1126/science.adr8382
  9. Caniels TG, Prabhakaran M, Ozorowski G, MacPhee KJ, Wu W, van der Straten K, et al. Precise targeting of HIV broadly neutralizing antibody precursors in humans. Science 2025;389:eadv5572. doi:10.1126/science.adv5572. https://doi.org/10.1126/science.adv5572
  10. Adepoju VA, Udah DC, Onyezue OI, Adnani QES, Jamil S, Bin Ali MN. Navigating the Complexities of HIV Vaccine Development: Lessons from the Mosaico Trial and Next-Generation Development Strategies. Vaccines 2025;13:274. doi:10.3390/vaccines13030274. https://doi.org/10.3390/vaccines13030274
  11. Varatharajan S, Krishnasai S, Girish C. Human immunodeficiency virus vaccines: Advances, challenges and future perspectives. World Journal of Virology 2026;15:116055. doi:10.5501/wjv.v15.i1.116055. https://doi.org/10.5501/wjv.v15.i1.116055