Shaping the Battlefield

A host of advances in vaccine design and evaluation promise to transform the campaign against HIV

By Regina McEnery

The cells of the adaptive immune response are, in many ways, the body’s elite forces. They may want for face paint and electronic gadgetry, but they patrol their biological perimeter as vigilantly as any contingent of commandos, gathering intelligence and adapting tirelessly to the ever-changing battlefield of the body. When they detect a threat, they respond swiftly, neutralizing turncoat cancer cells, viral saboteurs and invading microbial forces with exquisite precision and breathtaking ferocity.

Yet some enemies prove too canny even for these veterans of microscopic warfare. HIV certainly falls into that category. Not only does it specifically target a key officer in their command structure—the CD4+ T cell—but, thanks to its mutability, does so in ever-changing molecular disguises. When it is targeted by antibodies, the guided missiles of the immune response, it misdirects them with molecular decoys. To further complicate matters, the only neutralizing target open to attack is the Envelope protein, a complex of three identical pairs of proteins—the transmembrane gp41 protein and the extracellular gp120—that is essential to viral invasion of host cells.

But a string of scientific breakthroughs has lately exposed chinks in HIV’s formidable defenses. As was evident at the AIDS Vaccine 2012 conference in Boston, Sep. 9-12, researchers are today more confident than ever before that the field of HIV prevention is on the verge of a revolution. Several non-vaccine strategies—including microbicides and an assortment of treatment as prevention approaches— have lately shown promise in large-scale studies, setting the stage for a public health assault that could begin to turn the tide of the HIV pandemic.

The success of these approaches will certainly complicate vaccine development. Yet researchers remain optimistic that steady progress in a number of distinct but convergent strategies for HIV vaccine design and development could soon dramatically alter how we think about the future of the pandemic (see Q&A with Nelson Michael, this issue).

If any single aspect of that effort dominated the Boston conference, it was research into broadly neutralizing antibodies (bNAbs) against HIV, scores of which have been cloned and analyzed in recent years. Several laboratories, most notably at the Vaccine Research Center of the National Institute of Allergy and Infectious Diseases and within IAVI’s Neutralizing Antibody Consortium (NAC), have in recent years laid the groundwork to devise vaccine immunogens on the basis of information gleaned from their analysis.

The shape of things to come

William Schief, of the IAVI Neutralizing Antibody Center at The Scripps Research Institute in La Jolla, has been developing sophisticated computational methods to reconstruct the epitopes bound by bNAbs. At the Boston conference, he described how he and his colleagues obtained powerful proof of concept for this approach to vaccine design using as a model the respiratory syncytial virus (RSV), the most common cause of lung and airway infections in infants and toddlers. No preventive vaccine against RSV has yet been licensed, but a monoclonal antibody (mAb) called palivuzumab, approved by the US Food and Drug Administration in 1998, can block RSV infection when used to passively immunize premature infants and children with congenital heart disease. Manufactured by Maryland-based MedImmune, it targets an epitope on a protein that is essential to RSV’s invasion of host cells. The Schief group selected palivuzumab’s epitope for their immunogen design efforts.

Initial efforts to create an immunogen based on an X-ray crystallographic structure of that epitope in complex with motavizumab—an affinity-matured version of palivuzumab—obtained by Peter Kwong’s laboratory at the VRC proved disappointing. The first-generation experimental immunogens failed to elicit neutralizing antibodies in mice (1).

That might have been the end of it, but Schief’s laboratory didn’t want to give up. “We thought we could make better scaffolds,” Schief said, referring to the simulated protein structures that are engineered to hold a desired epitope in the appropriate conformation and orientation. Using new code written for Rosetta—a software suite that generates protein structures on the basis of peptide sequences—Schief’s lab generated from scratch a variety of possible structures for scaffolds into which the motavizumab epitope could be inserted. The software generated about 100,000 theoretical scaffolds. After filtering the results, the team ordered up eight genes, which were then expressed in Escherichia coli.

Six of the eight proteins, Schief said, were “well-behaved.” They were very stable and bound very tightly to motavizumab—much more so than structures generated in previous experiments. “The reason we got such tight binding,” says Schief, “is that we had frozen the conformation of the epitope.” Proof for that came from X-ray crystallographic structures of the unliganded epitope and one bound to motavizumab. They both fit almost perfectly when superimposed over a crystallographic structure of the antibody-bound conformation of the RSV peptide on which they were modeled. This established that the high affinity binding itself was real, not an artifact of aberrant contacts between the antibody and the scaffold.

To test its ability to elicit antibodies, Schief and his colleagues immunized rhesus macaques with various scaffolded epitope constructs and examined their sera for antibodies against RSV. Two assays, including a plaque reduction assay that is considered the gold standard for detecting and measuring the potency of neutralizing antibodies, found that by week 20 monkeys were making antibodies that neutralized a strain of laboratory-generated RSV that is highly resistant to neutralization. Vanderbilt Vaccine Center Director James Crowe, who ran the plaque assay, said the RSV strain neutralized in these assays measures up to strains found in the wild and has been used to challenge healthy adults in clinical trials.

Since macaques cannot be infected with RSV, Schief and his colleagues have no plans to pursue further studies in this animal model. But the preclinical success of the scaffolded epitopes establishes a significant proof-of-concept for an approach that Schief and many other researchers are now applying to design AIDS vaccines. “We were very excited because all of our work on scaffolds [designed on the basis of anti-HIV bNAbs] never produced any neutralizing antibodies,” said Schief. “And, you know, everyone rightfully would say, well, maybe scaffolding is never going to work. Maybe there are a lot of inherent problems with designing a little, minimal protein and eliciting antibodies to that that could cross-react to a big, complicated virus. But this experiment shows [our method] actually can work.”

A New Vibe in Boston

The conference in Boston might have been named AIDS Vaccine 2012, but three of its four opening plenary talks had little to do with vaccines. Instead, they covered vaginal microbicides, research into the early initiation of antiretroviral treatment to reduce transmission risk and an overview of what we’ve learned from trials of pre-exposure prophylaxis. The fourth talk, by Anthony Fauci, the executive director of the US National Institute of Allergy and Infectious Diseases (NIAID), focused on vaccines in the context of these and other preventive interventions. This underscored a larger point—that the new biomedical prevention strategies to prevent HIV are likely to change both the design and conduct of AIDS vaccine trials. It is likely to make vaccine development more expensive, if the availability of new prevention strategies drives down HIV incidence in cohorts and lengthens the timelines for trials. But it could also set the stage for real world evaluations of comprehensive prevention strategies that have the potential to reverse the tide of the AIDS pandemic. “The HIV prevention strategy will in fact be a unique paradigm of non-vaccine prevention modalities together with a safe and effective vaccine,” said Fauci. The field’s advocates have their work cut out for them, though. “We have gotten tired of making our own case,” said Bill Snow, executive director of the Global HIV Vaccine Enterprise, an organizer of the conference. “It sounds rote, obligatory and, worst of all, next to impossible. It needn’t be that way.” It is time, he argued, for the field’s advocates to “retune our rationale, refine our refrain. … We can’t promise a vaccine by a [certain date] but there are immediate, intermediate questions that will yield to intensive research.” This year’s conference was also notable for the prominent role given to young and early career investigators. Two of them, in fact, co-chaired the 2012 conference and about a third of the 46 plenary and symposium speakers were young and early career investigators. If the Boston conference was any indication, they’ll have no shortage of subjects on which to build a career. —RM

 

A minimalist approach

Structure-based design is not, by a long shot, the only approach to solving the neutralizing antibody problem. Some researchers are trying to strip the sole target on the viral surface—the HIV Envelope protein—down to its bare, immunogenic minimum.

Though essential to viral entry into target cells, the Envelope is a highly variable and structurally dynamic protein. Further, the functional protein offers up many non-neutralizing targets to antibodies, and so misdirects the response. It is covered with a coat of complex sugar chains that are identical to those found on human cells and are therefore ignored by the immune system. Those sugars also obstruct antibody access to potentially neutralizing epitopes on the underlying peptide.

To deal with these difficulties, researchers are using a variety of computational and protein engineering techniques to better focus the immune response on neutralizing targets on the Envelope trimer. Michael Cho, a biomedical science professor at Iowa State University, described in Boston how he and his colleagues stripped gp41 down to its extracellular domain, exposing the conserved membrane-proximal external region (MPER) that lies at the bottom of the intact HIV spike. The engineered protein, said Cho, bound tightly to a trio of bNAbs—2F5, 4E10 and z13e1—that are known to target the MPER domain.

Next, Cho used a histidine tag to link this gp41 fragment to zinc-chitosan, an adjuvant, and immunized rabbits subcutaneously four times over either 16 weeks or 24 weeks. Cho said eight of the nine rabbits mounted bNAb responses. Using the slower immunization regimen, neutralizing activity was observed against 42 viruses that spanned six different clades, including 27 relatively less sensitive Tier 2 viruses. One of the neutralizing epitopes overlapped those for 4E10, z13e1, and more recently identified 10E8.

Still, the lack of a precise structure for the functional trimer continues to impede its effective use as an immunogen. The closest scientists have come to actually viewing the unliganded form of the functional HIV Envelope glycoprotein has been with cryo-electron microscopy (cryo-EM), including the more refined single-particle cryo-EM (see IAVI Report blog, Aug. 20, 2012; A Slew of Science in Seattle, IAVI Report, Mar.-Apr. 2012). Cryo-EM involves snap-freezing the trimer in liquid nitrogen, taking its image from numerous angles, and then reconciling the images to reconstruct the structure.

A group led by Harvard structural biologist Bing Chen has used single particle cryo-EM to study gp140 trimers, and they plan to submit their structural analysis for publication soon. But Chen cautioned that cryo-EM has its limitations. It can produce inaccurate or misleading structures if the protein preparation being used contains a mixture of monomers, dimers and the like, rather than a homogeneous population of proteins. Chen described the difference between working with these kinds of mixtures as akin to taking 3,000 two-dimensional (2D) snapshots of an animal to reconstruct a three-dimensional (3D) model of the animal, or snapping 1,000 pictures each of a cat, monkey or dog to recreate a 3D model. The latter reconstruction, he noted, would “definitely not look like a dog or a cat or a monkey because the 2D images you used are not from the same object.”

To illustrate, Chen showed how the 3D structures of trimers made recently by different laboratories don’t all look the same. “At this point, you could not easily disregard one of these reconstructions as the wrong one,” he said. “They could be different conformational states. Or some are correct and some are not.” But such decisions, Chen pointed out, can have significant practical implications. “An incorrect EM structure of the Envelope trimer will mislead any effort to design or improve immunogens based on that particular structure.”

The long path to the bNAb

Even if scientists eventually compute and sculpt their way to the perfect or near-perfect Envelope immunogen, they still need to figure out how to get the body to make bNAbs. Trouble is, they are still trying to work out how exactly these antibodies are made. Studies that applied deep sequencing to trace the genetic pathways of bNAb maturation revealed that those that target the highly conserved CD4 binding site (CD4bs) of HIV, a common bNAb target, tend to share a genetic lineage. They also appear to have been refined over a long period of time, gathering large numbers of mutations in the process of affinity maturation. Many bNAbs also come from B cell lineages known for extremely long heavy chain complementarity determining region 3s (HCDR3s), which is a relatively rare trait.

A robust arsenal of bNAbs has already been isolated from chronically-infected individuals with HIV, and scientists are now studying other producers of these relatively rare antibodies for clues to eliciting bNAbs via vaccination. In one of the largest efforts to date, a study funded by the Center for HIV/AIDS Vaccine Immunology (CHAVI) analyzed the sera of 111 HIV-infected individuals from cohorts recruited by the Center for the AIDS Programme of Research in South Africa (CAPRISA ), CHAVI and Amsterdam to measure the prevalence of antibodies that target the CD4bs (2).

Barton Haynes, director of the Duke Human Vaccine Institute and leader of Duke CHAVI-ID—an offshoot of CHAVI—reported that the study found 88% of the samples contained antibodies that bind to the CD4bs. Some 47% contained antibodies to resurfaced stabilized core (RSC) probes, which preferentially bind bNAbs, such as VRC01. The data from each cohort differed some. The analysis found that 2-3 years after infection 31% of the CHAVI subjects and 32% of the CAPRISA subjects had made antibodies similar to those that typically target the CD4bs. But the percentage appears to rise over time: 79% in the Amsterdam cohort—a longer-studied group—were able to make such antibodies five years post-infection. Still, while the antibodies were present, sera from 23 individuals screened against a panel of six heterologous Env pseudoviruses found only modest neutralization potency and breadth, suggesting that the antibodies had not yet had the time to mature fully.

CHAVI-ID and the VRC are now using 454 sequencing to track the genetic lineage and maturation pathways of the antibodies found in the cohorts. They also hope to learn whether there are “blind alleys” in such processes that impede the ultimate generation of bNAbs. The ultimate goal, said Haynes, is to use this information to design immunogens that will guide and accelerate bNAb production.

The long arm of the bNAb

While the practical challenges thrown up by the circuitous pathways of bNAb affinity maturation worries researchers, Crowe thinks it may be possible to design immunogens that elicit high-affinity antibodies without having to go through such a drawn out process. Crowe’s work primarily focuses on the long HCDR3s common to HIV-neutralizing antibodies. The bNAbs PG9 and PG16, for example, both have extraordinarily long HCDR3 domains that are essential to epitope binding.

Crowe recently showed by comparing B cells in the peripheral blood at various stages of development that long HCDR3s appear not to be the products of somatic hypermutation (3). Three different subsets of B cells, including naive B cells and memory B cells, were isolated from the peripheral blood of four healthy HIV-uninfected individuals and four HIV-infected individuals, and their heavy-chain genes sequenced. Crowe and his colleagues reported that the naïve B cell subset encoded a higher proportion of antibodies with long HCDR3s than did affinity-matured B cell groups, suggesting that long HCDR3s are largely generated through recombination and the selection of unusual clones in the early phases of B cell development.

Crowe’s lab has also tried to identify antibodies in the B-cell repertoires of healthy donors that might have long HCDR3s that are predictive of the distinctive hammerhead shape of PG9’s, which it uses to interact simultaneously with glycans and the protein backbone of Env. To do this, they combed through six million sequences and found about 2,000 HCDR3s that looked to be the right length. Next, they took sequences of around 30 amino acids in length and forced them, in silico, to assume the shape of PG9’s hammerhead and asked the computer whether it liked what it had.

Only a few sequences were predictive of the hammerhead shape, and the one that Crowe’s lab thought would be most predictive ended up not being a good match after all. “But we have another antibody which the computer suggests has the [correct] shape and is able to maintain the shape in the context of a complex. We are predicting that this antibody from a naïve person would interact with HIV.”

Crowe said if scientists could figure out how to design an antigen that only binds long HDCR3s, but not the shorter ones, they might get around the problem of bNAb affinity maturation. “People are born with a repertoire that generates these long HCDR3s,” said Crowe. “It’s just that these [B cell clones] are rare.” This, he says, is cause for optimism. “The mystery is trying to stimulate that one in a million repertoire.”

Signatures of success

For all the momentum behind neutralizing antibody research, viable vaccine candidates based on bNAbs are not likely to reach clinical testing any time soon. In the meantime, scientists continue to uncover new clues about the only vaccine candidate that has demonstrated protection against HIV—the RV144 regimen, which was found to have 31% efficacy against HIV.

Researchers from multiple laboratories have been analyzing samples collected in the RV144 trial for insights into the underlying mechanisms of that protection. At last year’s AIDS Vaccine meeting in Bangkok, investigators shared the first set of results from such analyses, identifying what they called “correlates of risk” associated with the Thai regimen—a vCP1521 canarypox viral vector prime followed by a gp120 B/E AIDSVAX boost (4).

Those studies revealed, surprisingly, that one antibody response correlated with a reduced risk of HIV, while another correlated with an increased risk of infection (see A Bangkok Surprise, IAVI Report, Sep.-Oct. 2011). 
Scientists have since turned their attention to the antibody responses that correlated with a reduced risk of infection—namely, immunoglobulin G antibodies that bind to the V1/V2 region of HIV’s Envelope protein. 
Specifically, they have examined whether those vaccine-induced antibody responses selectively blocked certain HIV variants, and what genetic changes allow the virus to elude that targeting. Scientists refer to such escape as a “sieve effect.”

Led by researchers at the US Military HIV Research Program (MHRP), a key collaborator in the RV144 trial, the team examined nearly 1,000 HIV genetic sequences from 110 volunteers who became infected over the course of the RV144 trial—44 who received the candidate vaccine regimen and 66 who received a placebo. They then examined the viral sequences for evidence that the V2 region plays a major role in the modest protection seen in the trial.

Viruses that bore certain sequences in two stretches in the V2 region of the Envelope gene appeared to be vulnerable to vaccine-induced immune responses; viruses with mutations in those regions of the gene tended to evade such responses. One of the genetic signatures appeared to be associated with an efficacy as high as 78%. “This is an independent assessment that the V2 region is important,” said Morgane Rolland, lead author of the study and a scientist at MHRP.

The findings, published in the journal Nature the same day they were presented at the Boston conference, buttressed the credibility of the Thai trial results—adding to the molecular evidence that the observed protection was real and not just a statistical anomaly.

On the other hand, they underscored just how difficult it will be to design a broadly effective AIDS vaccine if all it takes to escape protection is a point mutation in a gene that is variable even by the standards of HIV. MHRP Director Nelson Michael acknowledged this fact, but was optimistic that the difficulties can be overcome. “We are making substantive progress in understanding what it will take to develop a more effective HIV vaccine, which will ultimately help us end this pandemic,” he said.

In a talk unrelated to the RV144 trial, Rolland reported results from a monkey study that found additional evidence supporting the importance of vaccine-induced responses against the V2 region on the simian immunodeficiency virus (SIV). The findings followed up on an earlier nonhuman primate study in which immunization of rhesus macaques with prime-boost regimens—an adenovirus serotype26 (Ad26) combined with modified vaccinia Ankara (MVA) or with MVA/Ad26—containing gag, pol and env genes from SIVsmE543 resulted in 80% or greater reduction in per-exposure probability of infection against a repeat intra-rectal SIVmac251 challenge. Because the challenge virus contained different viral sequences than those in the vaccine candidates, the results were particularly encouraging (5).

The follow-up study set out to determine if the observed protection was due to Env-specific antibody responses. To answer this question, 66 sequences from SIVmac251 challenge stock and 409 near-full length viral genomes from 13 vaccinated and 13 control monkeys were amplified and evaluated for evidence of a sieve effect. Rolland said that there appeared to be little overall difference in the full-length Env sequences in breakthrough viruses from either group of monkeys. But when they drilled down deeper, they did find evidence of a sieve effect in the Env-V2 segment, suggesting that antibody responses did play a role in the observed reduction in the risk of infection.

Another monkey study, led by Genoveffa Franchini, chief of the animal models and retroviral vaccine section at the US National Cancer Institute, found that a vaccine regimen against SIV analogous to the one used in the RV144 trial induced similar immune responses and outcomes. As was the case in the RV144 trial, the ALVAC-SIV/SIVgp120 prime-boost combination protected a third of the 75 Indian rhesus macaques challenged with a low dose of the highly-pathogenic SIVmac251 but did not slow disease progression in animals that were infected. The findings followed an earlier pilot study of 21 monkeys by Franchini’s lab that evaluated the same vaccine regimen and reached the same conclusions (see Tapping the Sanguine Humor, IAVI Report, Mar.-Apr. 2012.)

Harnessing Innate Immunity

A number of laboratories are developing HIV vaccine candidates that directly activate the innate immune response—especially by engaging dendritic cells, antigen presenting cells that have long been known to play a key role in vaccine-induced immunity. Preliminary results from a Phase 1 dose-escalation trial of one such vaccine candidate in 45 HIV-uninfected individuals, which was presented at the AIDS Vaccine Conference in Boston by Marina Caskey of Rockefeller University, suggest it is both safe and immunogenic. The vaccine candidate contains a monoclonal antibody engineered to recognize DEC-205, an endocytic protein found on the surface of dendritic cells that mediates efficient presentation of antigens. The antibody is fused to an HIV clade B p24 Gag protein. Gag p24 was picked as the candidate immunogen because it has many conserved epitopes, and is thus more likely to induce CD8+ T-cell responses against a broad range of viral variants. The vaccine candidate was administered at three different doses, along with an experimental adjuvant called Poly ICLC (Hiltonol) that is designed to stimulate innate immune responses. Volunteers received three subcutaneous vaccinations of either the vaccine candidate or a placebo over 12 weeks and were then monitored for 12 months. Though the study remains blinded, Gag p24-specific Immunoglobulin G (IgG) antibody responses were detectable in 60% of the volunteers in both low-dose and mid-dose groups at weeks 4,8,12 and 16. Interleukin-2 and tumor necrosis factor-α were the predominant cytokines detected. Antibody responses were durable, with titers unchanged six months after the last immunization. —RM

 

Stalled trials

Researchers hope to improve upon the results of RV144. But during a satellite session held prior to the opening ceremonies, Jerome Kim, deputy director of science at the MHRP, discussed issues impeding two such planned studies—one involving men who have sex with men (MSM) in Thailand and a second among heterosexual men and women in South Africa.

The Pox Protein Public-Private Partnership, or P5, launched a year ago to boost the vaccine efficacy seen in the RV144 trial to at least 50%, had hoped to launch both studies by late 2012. But a number of setbacks ranging from money to laboratory infrastructure to manufacturing have scuttled P5’s initial plans, said Kim. The earliest start date for the southern Africa trial is now pegged at late 2014, and it is unclear when the MSM trial in Thailand will get off the ground. “This is a little depressing,” Kim conceded.

The vaccine candidates slated to be tested a Phase IIb trials in Thailand and southern Africa contain immunogens specific to different HIV subtypes. But the candidates in both trials are delivered in a regimen that includes an ALVAC viral vector vaccine candidate as the prime, followed by a gp120 protein boost containing a well characterized adjuvant known as MF59. 

Kim said a major challenge in the MSM trial in Thailand has been finding a manufacturer for the proposed gp120 boost. The company that owns the intellectual property rights for the protein boost in the RV144 trial is unable to produce enough for another trial, which means a new manufacturer must be found. Novartis Vaccines and Biologics, in Cambridge, Mass, has the contract to make the protein boost for the southern Africa trial. Kim said Novartis has been asked to make protein for the Phase IIb trial in Thailand as well.

Patchwork protection

Another exciting vaccine strategy involves mosaic antigens—full-length or near-full length proteins that are created by stitching together genetic sequences that represent not only the broadest possible range of HIV variants but that have also been optimized for their potential to induce vigorous and effective immune responses.

Such mosaic immunogens have not yet been tested in clinical trials. But encouraging results from nonhuman primate studies have left researchers hopeful that the approach might be effective in both preventing HIV and in controlling viral replication among those who become infected despite vaccination.

It is, however, unclear how full length mosaic genes stack up against immunogens that encode strings of highly conserved fragments taken from various genes. It is at least possible that unconserved epitopes within full length mosaics will diminish the breadth and strength of responses to their conserved elements—which may be essential to improving the breadth of protection obtained against circulating HIV variants.

To test this notion, a team of researchers led by Dan Barouch at Harvard University compared the breadth and magnitude of the cellular responses induced in rhesus macaques who were immunized with either two or three full-length mosaic Gag, Pol, and Env immunogens, or mosaic immunogens stitched together from conserved regions of those genes. The recently published study used recombinant adenovirus (rAd26) prime in combination with a rAd35 boost to deliver the immunogens (6). “We thought the total magnitude of the cellular immune responses to [conserved elements of] the full-length immunogens would have been diminished, but we actually found the opposite,” said Kathryn Stephenson, a scientist in Barouch’s lab who presented results of the findings in Boston.

The full-length genes induced, as might be expected, a substantially greater breadth of HIV-specific cellular immune responses. But, contrary to the researchers’ expectations, these full length proteins also induced a greater magnitude and breadth of immune responses against conserved epitopes compared to the patchwork immunogens constructed only from conserved regions. The study also found that the breadth of cellular immune responses to the bivalent and trivalent immunizations was comparable. The study suggests not only that full length mosaics might be the better choice, but that simpler, bivalent immunogens work just as well as trivalent ones to elicit a broad response.

In another talk on mosaics, Bette Korber, who heads the HIV Database and Analysis Project at Los Alamos National Laboratory (see Tracking HIV Evolution, IAVI Report, May-June 2010), reported results from an ongoing immunization study in 36 rhesus macaques that induced remarkable resistance to a low dose, heterologous intrarectal challenge in vaccinated animals. The study is being led by Barouch and Michael of the MHRP, but Korber—whose database furnished the sequences used to build the full-length proteins expressing gag, pol and env genes—presented some of the recent findings during her plenary.

Korber reported that the vaccinated animals received various combinations of Ad26, MVA or Ad35 as either a prime or boost before being challenged with simian-human immunodeficiency virus (SHIV)162p3. Korber said the vaccinated animals were able to partially resist heterologous repetitive intrarectal challenge compared to the unvaccinated animals. She said their per-exposure risk turned out to be 80-90% lower than that of the sham-vaccinated controls. “These results are highly significant,” said Korber.

She also said the two mosaic inserts evaluated in the different viral vector combinations elicited strong CD4+ and CD8+ T-cell responses to Gag/Pol/Env, and good cross-clade binding antibody responses to Env. This animal study is very much a work in progress, with more data on what may be driving the protection expected soon. Korber did note, however, that an Ad26/MVA viral vector vaccine candidate bearing mosaic immunogens is being manufactured for evaluation in a clinical trial that researchers hope to begin soon.

1. J. Mol. Biol. 409, 853, 2011
2. J. Virol. 86, 7588, 2012
3. PLoS One 7, e36750, 2012
4. N. Engl. J. Med. 366, 1275, 2012
5. Nature 482, 89, 2012
6. J. Virol. 86, 11434, 2012