HIV vaccines Flashcards
successful antibody trial
RV144 trial, in which over 16,000 individuals were vaccinated with a prime-boost regime consisting of 4 injections of a replication-deficient canarypox-HIV vector (ALVAC-HIV) followed by 2 boosters of a recombinant gp120 subunit vaccine (AIDVAXXB/E).
• Primer = intended to stimulate cell-mediated immunity
• Boost = intended to stimulate humoral response (NA)
This vaccine is the most successful candidate vaccine to date, but still achieved only a modest efficacy of 60% at 12 months, and 31% at 42 months
secondary analysis of vaccine trial
. Secondary analysis by Haynes et al (2012) found that protection in this trial correlated best with antibody responses. In this study, they performed a range of cellular and antibody assays on samples collected from vaccinated individuals 2 weeks after their final vaccination, and found that IgG binding to gp120’s V1V2 loop correlated inversely with infection, with vaccine efficacy rising to 43% in reactive individuals. IgA against Env also key to protection.
In contrast, cytotoxic T lymphocyte (CTL) responses were not correlated with infection. Whilst only providing evidence of correlation and not causality, this study provides evidence that eliciting potent neutralising antibody responses will likely be the key to protection from HIV.
passive immunisation of blabs
One such study was conducted by Moldt et al (2012), who found that rhesus macaques could be protected from a high-dose challenge with a chimeric simian-HIV (SHIVSF162P3) by prior infusion of PGT121 bnAbs. They found that all 5 animals infused with doses of 1mg/kg or 5mg/kg of PGT121 displayed sterilising immunity, whilst 3/5 animals were protected with doses as low as 0.2mg/kg, which resulted in plasma titres of 1.8ug/ml deemed achievable via active vaccination. It is important to consider this study’s limitations; it only examines the protection provided by a single antibody against a specific viral variant, so we cannot know whether passive immunisation of PGT121 would provide sterilising immunity against a variety of heterologous isolates, and it is unclear whether the results would be as striking in humans as in the macaque model used. However, this study, particularly in the context of other similarly promising experiments, provides proof-of-concept that bnAbs can provide protection against HIV.
what are SOSIPs
A range of different Env-derived immunogens have been investigated as vaccines to elicit bnAbs, though stabilised native-like Env trimers, termed SOSIPs, have arguably shown the most promise. SOSIPs are good structural and antigenic mimics of Env gp140 trimers but contain a series of modifications that increase their stability above that of soluble WT Env, which quickly dissociates into gp120 and gp41ECTO when not contained within a viral membrane
This is a problem results in introduction of non-native, often immunodominant epitopes effectively distracts the immune system, making it unlikely that neutralising responses will prevail
These modifications include the insertion of a disulphide bond that covalently links gp120 and gp41ECTO, an I559P mutation in the gp41 subunit that stabilises the prefusion conformation of gp140, and truncation at position 664 to remove the MPER tail to reduce the likelihood of aggregate formation and increase solubility, and together encourage gp140 to maintain native-like conformations.
sosips vs WT env
SOSIPs are typically more immunogenic than peptide or monomeric gp120 immunogens as they facilitate the presentation of a wider range of epitopes, including quaternary and conformational epitopes bound by many bnAb populations. This was illustrated in a comparative study by Beddows et al (2007), in which rabbits were vaccinated with SOSIP.R6 gp140 or monomeric gp120 based on the Env sequence of the JR-FL neutralisation-resistant variant of HIV. They observed that a consistent neutralising response to HIVJR-FL was only generated in animals immunised with the SOSIP, indicating that SOSIPs are more likely to induce vaccine protection than monomers.
BG505 SOSIP.664
One SOSIP that has garnered significant attention is the BG505 SOSIP.664 gp140 trimer, examined in a study by Sanders et al (2013). This SOSIP is based on the amino acid sequence of Env from the subtype A BG505 strain, with modifications that increase stability and increase the number of bnAb epitopes. Biochemical analysis showed that these trimers resist dissociation and aggregation and are much more stable than corresponding gp120 monomers. When immobilised onto ELISA plates, the trimers were efficiently recognised by almost all bnAbs that neutralised the corresponding BG505 virus, including 2bnAbs (3BC315 + 3BC176) not bound by any other soluble Env protein ever tested = this indicates that epitopes on immunogen are intact and native-like. They also found that the bnAbs to quaternary-preferring epitopes generally interacted well with the SOSIPs, but poorly with monomeric gp120 or trimers lacking glycans, indicating that the soluble trimers maintain a more appropriate native-like conformation. In addition, the SOSIP trimers were bound poorly by non-neutralising antibodies, which bound well to gp120 monomers, suggesting non-neutralising epitopes are structurally occluded in the SOSIPs. This SOSIP has achieved promising results in animal testing elsewhere and entered phase I clinical trials (IAVI W001) in 2019.
nanoparticles study
Sliepen et al (2015), who produced ferritin nanoparticles capable of presenting 8 copies of BG505 SOSIP.664 and compared their immunogenicity to soluble SOSIP.664 trimers in mice and rabbits. They found that trimer binding responses were 8-fold higher in the mice and 2-3-fold higher in the rabbits immunised with nanoparticles than those immunised with soluble trimers. Furthermore, the rabbits vaccinated with nanoparticles exhibited higher neutralising antibody titres against tier 1A and B viruses than the other group, though this was not statistically significant for all viruses examined. This study provides evidence that nanoparticle display of SOSIP.664 will likely result in more potent antibody responses and greater neutralisation breadth than achieved with soluble trimers. However, it is important to note the small sample sizes used in this study (as few as 5 animals in each group). It is also unclear from this study how pronounced the difference in immunogenicity between nanoparticles and soluble trimers will be in humans, given the difference in trimer binding is much less significant in the rabbits than the mice.
what are nanoparticles
Research has indicated that the immunogenicity of SOSIP trimers could be enhanced by their incorporation into nanoparticles to permit multimeric antigen presentation. Presentation of repetitive Env trimers on nanoparticles helps to increase the avidity of low-affinity epitopes, increasing the likelihood that a complementary B cell is selected and survives in the germinal centre; as a result, nanoparticle vaccines can elicit neutralising antibodies against epitopes not elicited by trimeric antigen alone. Furthermore, nanoparticle presentation facilitates BCR cross-linking, resulting in improved B cell activation and increased affinity maturation, and shields the trimer base from recognition by non-neutralising antibodies. Disadvantage to method = nanoparticles themselves (e.g. ferritin) are often highly immunodominant
favouring recognition of immunorecessive epitopes
Ringe et al (2019), who redirected the neutralising antibody response against the BG505 SOSIPv4.1 trimer away from the autologous immunodominant 241/289 glycan hole epitope towards alternative sites, through the selective knockin and knockout of glycans. Whilst this strategy successfully induced the production of different neutralising antibodies at the expense of those targeting the immunodominant site, it disappointingly did not equate to an increase in the breadth of neutralisation against heterologous tier 2 viruses. However, this ‘closing and opening holes’ method could be promising and warrants further investigation = exploiting glycan immunorecessivness in both directions
what is GT?
Also being explored is a novel vaccine design strategy called germline targeting, in which individuals are vaccinated with immunogens that prime and stimulate specific B cells that produce the germline inferred variants of bnAbs. Iterative rounds of immunisation with different Env immunogens will hopefully facilitate affinity maturation of these cells, helping to shepherd them towards the production of mature bnAbs. It has been noted that germline precursors of bnAbs do not typically bind to wildtype Env proteins, so inducing their production requires the engineering of recombinant proteins with targeted mutations.
germline engineering study
Steichen et al (2019), which focused on inducing germline precursors of BG18 bnAbs that bind to the N332 supersite on gp120. They first used structural modelling and cryo-EM to deduce the structure of BG18’s binding mode, before screening a sequencing dataset of BCR heavy chains to identify rare B cell populations capable of producing BG18-like antibodies (e.g. BG18gH). From this, they engineered next-generation Envelope trimer immunogens, named N332-GT2, that displayed high affinity for these BCR chains. The N332-GT2 trimers were capable of activating rare germinal centre BG18 precursor B cells in a mouse model, not activated by reference trimers that didn’t possess germline-targeting mutations. Further experiments also indicated that immunisation with N332-GT2 helped induce affinity maturation towards bnAb production, with antibodies with high affinity for N332-GT2 extracted from the immunised mice at day 42 showing a 200-fold higher affinity for a more native-like Env trimer than naïve antibodies with affinity for N332-GT2.
problems with GT
This study, among others, solidifies germline-targeting as a key approach to consider in the development of sterilising vaccines. However, it is important to consider that this method will likely be limited by the fact that in humans, the frequency of B cell precursors that recognise germline-targeting immunogens is very low, and that germline BCR affinity for a bnAb epitope may also be very weak. As a result, individuals may require long-term exposure to a germline-targeting immunogen to facilitate the selection and activation of rare B cells. Furthermore, whilst studies such as this have resulted in the production of immunogens that can elicit germline precursors of bnAbs, comparatively little progress has been made in the development of subsequent immunogens required to shepherd the B cells through affinity maturation.
CTL approaches
- Naked DNA prime + Ad5 boost Gag, Pol, Env vaccine => SIV challenge => increased survival despite lack of protection from death; lower viral load in vaccinated macaques (Letvin et al. 2006)
- This led to The STEP and Phambili clinical trials = two phase IIb efficacy trials to test the effectiveness of Merck’s MRKAd5 HIV-1 gag/pol/nef trivalent vaccine. These recombinant vectors were selected for development by Merck due to positive results in rhesus monkey experiments. These two studies of 3000 individuals were halted early as the first analysis demonstrated a lack of efficacy and even indeed led to an increase in infection rates in the trial group in those who had pre-existing antibodies against the Ad5 vector. Still not sure what happened, but maybe Ad5 as a vector leads to CD4 T cell homing to mucosal surfaces – additional target cells, so increase in infection rates
obstacles to vaccine trials
Obstacles to vaccine trials
HIV incidence is less than 2 per-100 person years in target populations due to increased awareness of the disease and HIV treatments significantly reducing HIV positive individuals ability to transmit the infection. This necessitates very large studies and therefore very high costs to obtain data with enough statistical power to obtain usable results. As well as the nature of HIV infection with it having no curative treatments it isnt possible to rigorously test the vaccine as it would be unethical to expose individuals to HIV.
SIV vs SHIV trials of CTL vaccines
Shiver et al (2002) = showed that adenovirus type 5 (Ad5) vector vaccines, expressing SIV gag protein, protected against SHIV challenge in rhesus macaques
Induced potent epitope-specific CD8+ T cell responses, reduced viremia and CD4+ T cell loss, decreased/prevented disease progression
Shiver et al (2005) = Ad5 vaccine ineffective against intrarectal SIVmac239 challenge in macaques poor viremia control
Is SIV a more accurate model for evaluating vaccine efficacy?
Conclusions vaccines that protect macaques from specific SHIV challenge, but not pathogenic SIV are unlikely to provide protection against HIV in humans
