Aids Vaccine Researchers STEP Up to the Challenge

Questions linger, but so does determination, as researchers gather at the AIDS Vaccine 2008 Conference

By Kristen Jill Kresge

This year’s AIDS Vaccine Conference, which was held in Cape Town, South Africa, from October 13-16, was momentous on both political and scientific fronts. It was the first time the annual conference was held in an African country and Lynn Morris, conference chair and head of the AIDS unit at the National Institute for Communicable Diseases in Johannesburg, kicked off the conference by commenting on the particular significance of it being held in South Africa. “Nowhere else is the need for a vaccine greater than it is here,” she said, adding that this conference sent an important signal that “while we’ve suffered a setback, we’re not giving up.”

Even more politically significant were the remarks made by the newly appointed South African Minister of Health, Barbara Hogan. After just two weeks on the job, Hogan made one of her first public addresses to the nearly 1,000 conference delegates. “We know that HIV causes AIDS,” she said, immediately making her positions clear. “The science of HIV and AIDS is one of the most researched subjects in the medical field.” Hogan also praised the conference organizers for holding the meeting in South Africa. “The timing of this conference coincides with a renewed interest in HIV prevention in this country. To the South African government and its people, there can’t be any more important meeting to be held at this time.” She called for evidence-based public health education as well as the development of evidence-based HIV prevention tools, which she said were critical to changing the course of the epidemic, and confirmed South Africa’s commitment to conducting clinical trials of vaccines. Hogan’s comments stood in stark contrast to those of her predecessor and were lauded by subsequent speakers.

On the scientific front, this year’s meeting was momentous because it was the first to be held following the unexpected failure of Merck’s adenovirus serotype 5 (Ad5) vector-based vaccine candidate (MRKAd5) in the STEP trial last fall, just after the 2007 conference. Since then the landscape of the AIDS vaccine field has changed dramatically. “The whole meeting has been held in the fallout of the STEP trial,” said Edward Rybicki, a professor of microbiology at the University of Cape Town who served as a rapporteur on the topic of vaccine concepts and design. The conference provided an opportunity for researchers, clinical trial investigators, and advocates to get the latest data from the STEP and Phambili trials. “We watch with bated breath every new piece of data that comes out of that study,” said David Weiner, chair of the Gene Therapy and Vaccines program at the University of Pennsylvania School of Medicine. Researchers also discussed some of the lingering questions about the potential for cell-mediated immunity candidates and the value of non-human primate (NHP) models for predicting vaccine efficacy. Other areas of focus at the meeting included the role of innate immunity in HIV infection and the exploration of novel viral vectors.

Stanley Plotkin, executive advisor to the CEO of Sanofi Pasteur and veteran vaccinologist, was the final speaker at the opening session and his comments were echoed over the following days by many presenters (see An Interview with Stanley Plotkin). Plotkin summed up his views on the AIDS vaccine field with the expression ‘sang-froid,’ which means to avoid panic when things look bad. “While the situation is serious,” Plotkin said, “it’s not desperate.”

Emerging data

One of the key points of interest at the conference was, of course, the data emerging from the STEP trial. Since the results were first made public last September, they have practically become household news, at least in some circles. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases (NIAID) at the US National Institutes of Health (NIH), said during his keynote lecture, “even the gardeners at the NIH know the three [key] bullets of the STEP study.”

Julie McElrath, director of the Vaccine and Infectious Disease Institute at the Fred Hutchinson Cancer Research Center in Seattle, outlined progress in analyzing the data from the STEP trial in a plenary talk. She first noted that only 31% of vaccinees in the STEP trial mounted both CD4+ and CD8+ HIV-specific T-cell responses following three vaccinations with MRKAd5. In his earlier comments, Plotkin said only 0.5-1% of total CD8+ T cells in vaccinees were specific to HIV, suggesting to him that the candidate’s failure could be due to the paucity of immune responses it induced. “The responses were inadequate,” said Plotkin. In fact, when McElrath and colleagues compared the magnitude and breadth of the T-cell responses induced by MRKAd5 to those observed in a group of long-term nonprogressors (LTNPs)—individuals infected with HIV who are able to control viral replication or disease progression for an extended period of time without the aid of antiretroviral therapy—they found the median percentage of vaccine-induced CD8+ T-cell responses in STEP volunteers were 43% lower than the average in HIV-infected LTNPs. “If we’re trying to mimic the responses in these individuals, we’re not there,” said McElrath.

However, Bruce Walker, director of the Partners AIDS Research Center at Massachusetts General Hospital, continues to find that individuals in his cohort of elite controllers—a subset of LTNPs who maintain viral loads less than 50 copies of HIV/ml of plasma—have weaker CD8+ T-cell responses than individuals with typical HIV disease progression. One key difference he’s detected so far among elite controllers is that their CD8+ T-cell responses are directed primarily toward Gag. Among STEP trial volunteers, McElrath noted that vaccinees without pre-existing Ad5 immunity were “more likely to make a Gag response,” and in individuals with higher levels of Gag-specific T-cell responses, there was a trend toward lower viral load. According to McElrath, there was also an inverse correlation in vaccine recipients, without pre-existing Ad5 immunity, who subsequently became HIV infected between the level of interferon (IFN)-γ-secreting T cells directed toward HIV Gag and viral load. She referred to these findings as “some potential ray of hope” but also acknowledged that there are “a lot of caveats associated with this,” including the very small number of volunteers that these analyses are based on.

Still, McElrath said these preliminary findings suggest that a CD8+ T-cell response capable of reducing viremia may be an attainable goal. However when McElrath and colleagues analyzed the function of the CD8+ T cells induced by a similar Ad5 vaccine candidate—in this case the Ad5-vector based vaccine candidate developed by researchers at the Vaccine Research Center (VRC) at NIAID—using a viral inhibition assay, the results were not encouraging. Peripheral blood mononuclear cells (PBMCs) collected from two volunteers in a Phase I trial who received two doses of the VRC’s Ad5 candidate were evaluated. In this assay, CD4+ and CD8+ T cells were isolated from PBMCs. The CD4+ T cells were infected with HIV in vitro, and the CD8+ T cells were then added to see if they had any effect on HIV replication. The CD8+ T cells from these individuals had no effect on suppression of HIV replication in this assay. McElrath said the caveat with this data is that the cell samples tested were collected at 52 weeks, a time when they were “less likely to be effector cells.”

Analysis of granzyme B and perforin expression, two proteins that induce apoptosis of virus-infected cells, are also being conducted with ex vivo CD8+ T cells from STEP trial volunteers and McElrath says this information may help “tease out what would be a more effective immune response against HIV.” Susan Buchbinder, principal investigator of the STEP trial and a professor of medicine at the University of California in San Francisco, said in another plenary talk that there was also a “tantalizing hint” that protective human leukocyte antigen (HLA) types were associated with a reduction in viral load among vaccinated volunteers, but this is a very preliminary finding based on only a small number of participants.

In the meantime, researchers are still looking for any effect MRKAd5 may have had on HIV progression in vaccinated volunteers. Holly Janes, an assistant member of the biostatistics program at Fred Hutchinson Cancer Research Center (FHCRC), presented data in the late-breaker session on a small sub-group of HIV-infected male volunteers from the STEP trial—33 who received placebo and 40 who received MRKAd5. Of these volunteers, 25 have already initiated antiretroviral therapy. Janes reported that there was no significant difference between the median viral load in vaccine and placebo recipients prior to starting therapy, nor was there a difference in the amount of time prior to their initiation of treatment. The pre-treatment CD4+ T-cell counts and set-point viral loads were also similar between vaccine and placebo recipients. Janes said the data did not provide any evidence to suggest that the vaccine had exacerbated HIV disease progression.

In another study, researchers investigated a possible correlation between levels of pre-vaccination Ad5 neutralizing antibodies and the induction of Ad5-specific T-cell responses in STEP trial volunteers. High Ad5 antibody levels was one factor associated with an increased risk of acquisition of HIV among vaccine recipients in the STEP trial and one possible explanation for this was that individuals with higher Ad5 antibody levels would also have higher levels of Ad5-specific memory CD4+ T cells that would become activated upon vaccination, creating more target cells for HIV. However, according to research presented by Nicole Frahm, a faculty member in infectious diseases at FHCRC, individuals in the STEP trial with high Ad5 neutralizing antibody titers actually had lower amounts of Ad5-specific CD4+ and CD8+ T cells. “Ad5 [antibody] titers don’t really tell you if someone will make Ad5-specific cellular responses or not,” said Frahm.

She and her colleagues incubated PBMCs from 139 volunteers in the STEP trial who received three doses of MRKAd5 with 10,000 empty Ad5 vector particles per cell and then analyzed the samples for Ad5-specific T cells using intracellular cytokine staining. This study was conducted with the empty Ad5 vector developed at the VRC, but researchers first verified that it produced the same results in their assays as Merck’s Ad5 vector.

Frahm reported that 80% or more of vaccinees mounted Ad5-specific CD4+ T-cell responses and 65% or more had Ad5-specific CD8+ T-cell responses. The response rates varied based on Ad5 antibody titer, but both CD4+ and CD8+ T-cell response rates to the Ad5 vector were consistently higher in the Ad5 seronegative group. And when researchers compared Ad5-specific T-cell response rates from individuals who subsequently became HIV infected with those who did not, they found a significantly higher Ad5-specific CD4+ T-cell response rate in individuals who did not eventually acquire HIV, indicating that Ad5-specific T cells, at least in peripheral blood, were likely not responsible for the increased risk of HIV infection among vaccinees. “This was not what we expected at all,” said Frahm, who cautioned that researchers still need to look at the level of Ad5-specific T cells at mucosal sites. “We could just be looking at the wrong spot,” she warned.

Dan Barouch, associate professor of medicine at Beth Israel Deaconess Medical Center in Boston, reached a similar conclusion after analyzing samples from 116 individuals who participated in a Phase I trial of an earlier version of MRKAd5 encoding only Gag. He found that individuals with high levels of pre-existing Ad5 antibodies had higher Ad5-specific antibody responses following vaccination, but not higher levels of Ad5-specific T-cell responses. He is planning NHP studies to see if Ad5-specific T cells are concentrated at the mucosa.

While the reasons for MRKAd5’s failure are still unknown, and may never be completely clear, McElrath said investigators affiliated with the STEP trial have “made further progress in defining threshold responses for T-cell based vaccines.”

Phambili data

Glenda Gray, executive director of the Perinatal HIV Research Unit in Soweto, South Africa, presented data collected so far from the Phambili study, the second Phase IIb study of MRKAd5, which is being conducted in South Africa. When immunizations in the Phambili trial were stopped last September, 801 volunteers had been enrolled, and 50% were women. While the STEP trial volunteers were primarily men who have sex with men, the intention in the Phambili trial was to evaluate the efficacy of the candidate against primarily heterosexual HIV transmission.

Of the 400 volunteers in the vaccine group when immunizations were ceased, 66% had received two vaccinations and 7% had received all three. Gray reported that so far there have been 29 HIV infections among the 801 volunteers, 17 within vaccine recipients and 12 in the placebo group. Most of these infections were in volunteers with pre-existing immunity to the Ad5 vector, as was the case in the STEP trial, with 16 of 17 infections in the vaccine group and 9 of 12 infections in the placebo group having occurred in individuals with measurable Ad5 antibody titers. Of the seven infections that have occurred among male volunteers in the Phambili trial, six were in uncircumcised men—four in the vaccine group and two in the placebo arm—another risk factor associated with HIV acquisition in the STEP trial.

Gray noted that the unblinding of volunteers in the Phambili trial has had a significant impact on HIV acquisition rates in the study—since the volunteers were told a year ago whether they received vaccine or placebo, no new HIV infections have occurred among vaccinated volunteers. “Unblinding was a major confounder,” Gray said. Because of this, she declined to make any comparisons between the Phambili data and the results of the STEP trial, which suggested that the vaccine candidate may have increased the risk of acquisition of HIV infection in certain subsets of volunteers, primarily uncircumcised men with high levels of pre-existing Ad5 immunity.

Debating the way forward

At this year’s conference there were two organized debate sessions at which pairs of researchers faced off over central questions currently dominating discussion in the AIDS vaccine field. These debates were peppered with references to the US presidential race and often took amusing side-turns as researchers built their arguments by gently provoking their opponents. It was an opportunity for researchers to display their creativity outside the lab, and video clips as well as doctored photos were used to great effect.

The first debate was about whether the NHP model should be used as a gatekeeper to clinical trials of vaccine candidates, or more specifically, whether a candidate vaccine should be required to show efficacy in NHPs to be advanced into Phase I clinical trials. And although there is general agreement that studies in NHPs play an important role in vaccine research, there is some disagreement about just how heavily the field should rely on them.

The utility of the NHP model in preclinical evaluation of vaccine candidates was one theme that emerged from the NIAID-sponsored HIV Vaccine Summit, held earlier this year to discuss future research priorities in light of the STEP trial (see Balancing AIDS vaccine research, IAVI Report, March-April 2008). In studies with rhesus macaques, MRKAd5 modulated viral load against challenge with a hybrid SIV/HIV or SHIV but did not provide any protection against a more stringent simian immunodeficiency virus (SIV) challenge (see Getting it right early, IAVI Report, Sep.-Dec. 2007). This was enough to convince some researchers that protection against SHIV was not sufficient to predict efficacy in humans and as Jeff Lifson, head of the retroviral pathogenesis section at the National Cancer Institute said, “SHIVs are [now] out of fashion.” But Lifson was quick to point out that “there is not a primate model, there are many primate models,” and he said it was important to understand the different models and use them thoughtfully so they can best inform research.

Paul Johnson, associate professor at Harvard Medical School, said he wanted to dismiss at the outset the idea that monkeys lie, an oft-repeated line in the field. “As long as we ask them the right questions they tell the truth,” Johnson said. “We are fortunate indeed to have a very robust animal model.” He and Lifson argued that to gain entry into Phase I/II trials vaccine candidates should have to show significant immunogenicity, which they defined as greater than a 1.0 log copies/ml difference in viral load in vaccinated NHPs as compared to controls, following homologous SIV challenge. To warrant testing in a Phase IIb, screening-test-of-concept (STOC), or Phase III trial, they suggested a candidate should have to provide improved protection against homologous challenge when compared to MRKAd5 or provide protection against heterologous SIV challenge. Johnson and Lifson said criteria other than protection data in NHPs could argue for testing some candidates, including those that elicit neutralizing antibodies. In the case of candidates based on viral vectors with limited replicative capacity in macaques, they said the onus would be on the developers to decide when it was appropriate to advance a candidate into humans.

Johnson said the finite financial, manufacturing, and human resources, along with the need for extensive iterative cycles of testing AIDS vaccine candidates, argued for such guidelines, citing several examples of vaccines that were immunogenic in monkey models and then in humans. He said the positive predictive value of the NHP model could only be proven when a vaccine also showed efficacy in humans, but results from both the STEP trial and the Phase III trial previously conducted by VaxGen have paralleled the results seen in NHP studies.

Weiner and Jerald Sadoff, chief executive officer of the Aeras Global TB Vaccine Foundation, agreed that any vaccine platform should be studied in relevant macaque challenge models, but they argued that protection in NHP studies should not be required prior to advancing a candidate into Phase I trials. “It doesn’t matter what works in monkeys,” said Sadoff, “it only matters what works in humans.” He cited several vaccines, including those against cholera and rotavirus, which were developed without the use of relevant animal models as well as others like the malaria vaccine that is currently in efficacy trials, which were thrown off track by the data collected in NHP studies. “In malaria we were completely misled by the monkey model,” he said.

He also argued that although SIV is analogous to HIV, it’s not the same. “We have a different physiology in the animal and a different pathogen,” added Sadoff. He and Weiner concluded that while monkey models should be used as an immunogenicity marker for Phase I trials, they should not serve as a gatekeeper. Rather, Sadoff suggested that NHP studies should be done in parallel with Phase I trials so that clinical evaluation isn’t delayed.

Alan Bernstein, executive director of the Global HIV Vaccine Enterprise, said linking NHP researchers with clinical trial researchers was one project the Enterprise will be spearheading in the coming months. “We need to regard clinical trials as science,” he said, adding that the separation of clinical and discovery research “is a false dichotomy.” The Division of AIDS at NIAID will hold an NHP workshop November 12-13, 2008, at which experts on NHP research will offer guidance on how NIAID should invest in NHP research to advance the discovery of an AIDS vaccine.

Size matters

The second debate session was focused on whether additional cell-mediated immunity (CMI) candidates should be advanced into efficacy trials, given the failure of MRKAd5. In this session Gary Nabel, director of the VRC at NIAID, and David Watkins, a professor at the University of Wisconsin-Madison, squared off against Dennis Burton, a professor of immunology at The Scripps Research Institute in California and scientific director of the recently established HIV Neutralizing Antibody Center (see Vaccine Briefs, this issue). Robert Gallo, director of the Institute of Human Virology in Maryland, was scheduled to be Burton’s debating partner but he did not attend the meeting.

This debate was originally scheduled when Fauci was still considering whether NIAID would fund a Phase IIb trial to test the VRC’s DNA/Ad5 candidates, according to John Moore, professor of microbiology and immunology at Weill Cornell Medical College, who served as moderator. That trial was initially postponed after the STEP trial results were released, and in July Fauci rejected the proposed Phase IIb trial design, known as PAVE 100A (seePAVEing the way to a smaller trial, IAVI Report, July-Aug. 2008). NIAID is still considering conducting a smaller trial to look only at the ability of the prime-boost DNA/Ad5 regimen to lower viral load in vaccinated individuals that become HIV infected, and the protocol for such a trial is still under development.

Nabel argued that efficacy trials of T-cell vaccines should continue and that these trials should be “sufficiently large to be able to address questions related to immune correlates, viral load, and prevention of infection.” But Burton said there are “too many uncertainties at this time” to justify large-scale trials. He said large efficacy trials would only distract researchers from the best-case scenario for an effective AIDS vaccine—one that stimulates both arms of the immune system. However, Watkins said testing CMI vaccines was essential since “we don’t have any candidate antibody-based vaccines yet.”

Nabel cited the increasing evidence of the efficacy of T-cell candidates in relevant nonhuman primate challenge studies as another reason to move forward with large-scale trials. Some of this data stems from studies by Watkins and colleagues and in a late-breaker talk, Nancy Wilson, an associate scientist in Watkins’s lab, presented data from a study in which rhesus macaques were vaccinated with a DNA/Ad5 regimen encoding all of the SIVmac239 genes except env. While vaccinees in the STEP trial developed immune responses to three to five HIV epitopes on average, the vaccinated macaques in this study developed immune responses to an average of 20 SIV epitopes. Following five low-dose, mucosal challenges with the heterologous swarm virus SIVsmE660, five of the eight vaccinated macaques were SIV infected. The vaccinated macaques had markedly lower viral loads—the average peak plasma viral load was 12,600 copies/ml, compared to four million copies/ml in unvaccinated control animals. And at eight weeks post-infection, the average viral load in vaccinated animals was undetectable, while the average in unvaccinated controls was 200,000 copies/ml.

On this point Burton didn’t disagree. “I have no problem with screening-test-of-concept trials with T-cell based vaccines,” said Burton, “particularly ones that show robust responses in the macaque model.” He voiced support for smaller studies, like STOC trials, that would involve fewer volunteers but could provide preliminary information about the ability of such candidates to lower viral load. It seemed that the division between the two sides in this debate revolved mainly around the use of the words ‘large scale.’ “For those of you who are looking for a fight,” Burton said, “I’m afraid you’re not going to get it.”

Better vectors?

The pool of CMI candidates that may be up for testing in future clinical trials will likely be based on novel vectors. One of these is the cytomegalovirus (CMV), which is under investigation by Louis Picker, associate director of the vaccine program at the Vaccine and Gene Therapy Institute at the Oregon Health and Science University. Picker said vaccines that generate typical T-cell memory responses, which are primarily central memory T cells (Tcm), may not be able to overcome what he called the “kinetic mismatch” between the explosive replication capacity of HIV and the ability of Tcm to expand, differentiate, and migrate to the sites of virus replication. Picker said most prime-boost vaccine regimens currently being tested induce mostly Tcm, whereas live-attenuated SIV vaccines, which provide the best protection seen so far in NHP models, induce mostly effector memory T cells (Tem). He therefore proposes that a vaccine candidate that could induce more Tem than Tcm may increase the potential for protection.

To study this hypothesis, Picker chose to evaluate a CMV-vector based vaccine, which he called the “quintessential inducer of effector memory dominant T-cell responses.” Picker vaccinated 12 macaques with a rhesus CMV vaccine encoding SIV gag, rev, nef, tat, and env and then repeatedly challenged them with low-dose SIVmac239. The vaccinated animals required a median of eight doses of challenge virus to develop a progressive infection, compared to only two doses in the 16 unvaccinated control monkeys.

Remarkably, 4 of the 12 vaccinated animals resisted progressive SIV infection altogether, though they were demonstrably infected by the challenge virus because they developed de novo immune responses to SIV antigens pol and vif, which were not included in the vaccine. Two of these four animals had transient, very low plasma virus levels, while the other two had no detectable virus in their plasma at all. There was still no sign of viral replication even after depleting the CD8+ T cells of these monkeys 133 days post challenge. Picker said this suggests that early Tem recognition of viral antigens may lead to an effector response that is capable of controlling viral replication early, likely at the sites of viral entry. Vaccine-elicited effector T-cell responses may need to be induced at sites of viral entry to control replication of the founder population of virus before a systemic infection is established, according to Picker. He said this also emphasizes the importance of conducting such low-dose mucosal challenge studies with vaccine candidates.

Another vector under development is the measles virus. Hussein Naim, director of vaccine research at the biotechnology company Crucell Berna Biotech in Switzerland, presented data on a replication-competent measles virus vector his company is developing based on a commercially-available live-attenuated measles vaccine. The measles vaccine, which is administered to children as part of a combination vaccine that also protects against mumps and rubella, provides long-lasting protection against the disease and has reduced morbidity associated with the virus by between 95% and 100%.

One advantage of using the paramyxovirus that causes measles as a vector is its ability to infect numerous types of cells, including macrophages, dendritic cells, lymphocytes, and monocytes. It also has a favorable safety profile, since it has been administered to millions of people, and is inexpensive to manufacture. The potential disadvantage to using measles as a vector is pre-existing immunity—measles vaccination is nearly universal among infants in developed countries. In studies with mice, passive administration of measles antibodies did somewhat inhibit induction of cellular and humoral immune responses following immunization with a measles vector encoding HIV antigens, according to Naim, but he said the vaccine candidate was still able to induce HIV-specific immune responses. Naim and his colleagues are now considering evaluating intra-nasal administration of an aerosol measles vector, in collaboration with researchers at NIAID, to further evaluate its ability to induce immune responses against HIV. Researchers at the Institut Pasteur in Paris are also collaborating with GlaxoSmithKline Biological to develop a measles vector.

Meanwhile Barouch and colleagues have initiated a Phase I trial with an adenovirus serotype 26 (Ad26) vector, which in a prime-boost combination with Ad5 provided better results than a heterologous Ad35/Ad5 combination in NHP studies. The Ad26/Ad5 prime-boost combination resulted in a 1.4 log copies/ml reduction of peak viral load and a 2.4 log copies/ml reduction in viral load set point in rhesus macaques, which persisted for more than 500 days following SIVmac251 challenge. Barouch observed that the correlate of viral control with this regimen was anamnestic SIV Gag-specific responses, which he said “looks similar to Bruce Walker’s data from elite controllers.”

DNA 2.0

Other researchers are continuing to explore different formulations and administration techniques to enhance the immune response profiles induced by DNA vaccine candidates. “DNA vaccines are improving rapidly,” said George Pavlakis, head of the Human Retrovirus Section at the National Cancer Institute. His group found that using different HIV antigens affects the immune responses induced by their DNA construct. Pavlakis and colleagues also found additional benefit in terms of immunogenicity using in vivo electroporation, a vaccine administration technique that uses electric pulses to disturb cellular membranes and thereby allows molecules like DNA to cross into cells more easily. Vaccinating rhesus macaques by electroporation with DNA vectors encoding the majority of SIVmac239 proteins, together with interleukin (IL)-12 DNA as an adjuvant, increased the level of antigen expression, improved cellular (central memory CD4+ T cells and effector CD8+ T cells) and humoral immune responses, and resulted in an improved level of protection following SIV challenge. The SIV-specific T-cell responses were able to secrete multiple cytokines, including IFN-γ, IL-2, and TNF-α, in response to SIV peptides and lymphocytes recovered from bronchioalveolar lavage showed that there was migration of SIV-specific memory T cells to this peripheral mucosal site. These immune responses were long-lived in the animals—they could still be detected in both blood and mucosa 10 months after vaccination.

When vaccinated animals were challenged mucosally with SIVmac251, they had greater than 10-fold lower viral loads in comparison to unvaccinated control animals. Pavlakis said IL-12 continues to be a good adjuvant for DNA vaccines, even with electroporation, and that this data shows that DNA vaccines can control viremia following highly pathogenic SIV challenge. He said this DNA construct is “as good, or as bad, depending on your perspective, as any other vaccine modality save for live-attenuated [SIV] vaccines.”

Innate immunity in vertical transmission

As researchers continue to develop vaccine approaches that could elicit antibody and cellular immunity, a growing level of interest is also being paid to the role of innate immunity in HIV infection. “Innate immune responses are clearly valuable,” said Plotkin. Both he and Fauci mentioned the importance of exploring the role of innate immunity, and of natural killer (NK) cells—a major component of the innate immune system, which also play a role in adaptive immunity—in particular.

In a late-breaker talk, Caroline Tiemessen, a professor of virology at the University of the Witwatersrand in Johannesburg, presented intriguing data on the role of NK cells in preventing vertical transmission from HIV-infected mothers to their infants. Tiemessen and colleagues looked at CD4+ and CD8+ T cells, as well as several subsets of CD3- cells, in 79 HIV-infected mothers and their 76 infants and found that CD3- NK cells that respond specifically to HIV peptides were associated with protection against vertical transmission.

According to Tiemessen, 43% of mothers and 16% of infants had CD3- responses specific to Env, and 22% of mothers and 5% of infants had CD3- responses specific to Reg [Tat, Rev, Vif, Vpu, and Vpr combined]. Most of the infants with HIV-specific CD3- NK-cell responses were born to mothers who also had high levels of these cells, and in mothers they were associated with lower viral loads and higher CD4+ T-cell counts, two factors that reduce the likelihood of vertical HIV transmission. Tiemessen also observed that HIV-specific NK cells were also dependent on a soluble plasma factor, as well as interactions with CD4+ T cells and HLA Type I molecules.

NK cells, which were assumed to non-specifically control virus, may “see” HIV in a specific manner and also play a role in ADCC, according to Galit Alter, an instructor at the Partners AIDS Research Center in Boston, who served as conference rapporteur on the topics of T-cell immunology and innate immunity. Unraveling the precise role of NK cells may lead to new avenues in vaccine research; then researchers will have to figure out how to harness the innate immune system, according to Fauci. “This is a scientific problem and there will be a scientific solution,” he said, “I believe we are on our way toward that goal.”