CROI Covers Advancements from Start to Finish

Highlights of recent HIV meeting run gamut from basic science to HIV prevention and vaccine research

By Kristen Jill Kresge

The 13th Conference on Retroviruses and Opportunistic Infections (CROI), which took place from 5-8 February, often struck a historic chord as many plenary and keynote speakers acknowledged the passage of important landmarks in the battle against the AIDS pandemic. The Annual Bernard Fields Memorial Lecture given by Bette Korber of Los Alamos National Laboratories during the opening ceremony helped set this tone. Korber began by explaining the genetic diversity of HIV, gazing all the way back to the 1930s when HIV first entered humans. This historical theme continued with other researchers who stopped to review and reflect on the progress of the field spanning relatively shorter time frames, including the 25 years since the first AIDS cases were reported in the US and almost a decade since the advent of highly active antiretroviral therapy (HAART).

But the almost 4000 researchers convened in Denver took away more than a history lesson. Presentations at the meeting also covered the most recent advancements in understanding the immunological events that occur early in HIV infection, using new strategies to prevent transmission of the virus, and vaccine vector development and studies that may help to elucidate the immune responses required to prevent infection. Mario Stevenson of the University of Massachusetts, a member of the conference committee, reminded the audience that these developments in both treatment and prevention remain an imperative since despite 25 years of research and innovation, not a single HIV-infected individual has been able to successfully clear the infection and an effective vaccine hasn't yet been found.

Beyond a gut feeling

A highlight from the basic science track at the conference was the expanding body of knowledge on the immunological events that occur in early SIV/HIV infection in animal models and humans. Following on publications from Mario Roederer of the Vaccine Research Center (VRC) and Ashley Haase at the University of Minnesota last year (Nature 434, 1148, 2005; Nature 434, 1093, 2005), presentations at CROI contributed further evidence for the critical depletion of memory CD4+ T cells in mucosal tissues during acute SIV infection in several species of primates and added some new theories on the complicated dynamics of depletion and subsequent restoration of these cells during HAART.

The original work by Roederer and Haase showed that a decisive and rapid depletion of this subset of immune cells in the gut-associated lymphoid tissue (GALT) occurred within days of SIV infection in rhesus macaques (seeResearch Briefs and Research at the Extremes, IAVI Report  9, 2, 2005). This massive and very early immunological destruction in the GALT, and other mucosal tissues where large numbers of CD4+ T cells are found, probably plays a key role in the eventual immunodeficiency of these SIV-infected animals. Protecting these cells during acute infection has therefore become of increasing interest to both clinicians and vaccine scientists.

Two presentations at CROI focused on the events of acute infection in two other primate models, African green monkeys and sooty mangabeys, which unlike macaques are natural hosts for SIV and don't progress to AIDS after infection. This makes them an interesting model for studying any possible connection between early immunological events and disease progression.

Ivona Pandrea from the Tulane National Primate Research Center in Louisiana looked at the loss of CD4+ T cells in GALT during SIV infection in African green monkeys (Abstract 37www.retroconference.org/2006). Even prior to infection, these monkeys have fewer CD4+ T cells expressing the CCR5 receptor, which were identified by Ron Veazey, also of the Tulane National Primate Research Center, as the target cells for SIV infection in the macaque model. Pandrea compared the CD4+ T cell depletion in six African green monkeys and four rhesus macaques infected with SIVagm (African green monkey SIV).

What Pandrea observed was that both African green monkeys and macaques had very similar responses to infection in peripheral blood and in the GALT during chronic infection. Starting at day 28, Pandrea reported a dramatic depletion of these cells in both species and after 60 days, there were no measurable CD4+ T cells in the GALT of the African green monkeys. These cells recovered slightly after two years, but still remained lower than pre-infection levels. However the extensive damage that occurred in the GALT of African green monkeys did not have any pathogenic consequences and, despite a sustained high level of viral replication, none of these monkeys showed signs of disease progression.

But other groups have postulated that immune activation is also necessary for progression to AIDS and disease-related morbidity and could play a role in preventing the restoration of the immune system (Nat. Immunol. 7, 235, 2006). To address this Shari Gordon, who works in the laboratory of Guido Silvestri at the Emory Vaccine Center in Atlanta and her collaborators looked at the depletion of CD4+ T cells as well as the level of immune activation in the mucosal tissue of sooty mangabeys, another species that does not progress to AIDS when infected with SIV (Abstract 36).

Previous studies from Silvestri's group have shown SIV-infected sooty mangabeys maintain low levels of immune activation, as measured by T cell proliferation, despite high viral replication (J. Virol. 79, 4043, 2005). Gordon's presentation at CROI centered on an experiment comparing the depletion of CD4+ T cells in the peripheral blood, lymph nodes, and mucosal-associated lymphoid tissue (MALT) of 18 SIVsm-infected sooty mangabeys and 10 control animals. Samples of mucosal tissues were obtained by rectal biopsy and broncho-alveolar lavage.

Gordon observed that the SIV-infected animals had a significant depletion of CD4+ T cells in both the rectal and lung samples as compared to the uninfected controls, although the level of damage varied between animals. This reduction in CD4+ T cells in the MALT occurred despite overall low levels of immune activation, including activated or proliferating T and effector cells, measured in the mucosal tissues. Even though the CD4+ T cells were significantly depleted, in the absence of immune activation, these animals still did not progress to AIDS. Gordon and her colleagues are now looking for a possible correlation between the level of cell depletion and the degree of immune activation in sooty mangabeys.

"Ultimately, the goal of studying non-pathogenic SIV infection is to understand the mechanisms enabling these animals to avoid AIDS despite levels of viral replication that would be highly pathogenic in HIV-infected humans," says Gordon.

Another area of interest among researchers is the ability of both monkeys and humans to undo the damage wrought on these critical immune cells during acute infection. Several studies at CROI looked at how the introduction and time on HAART influences the restoration of the CD4+ T cells during chronic infection.

Gordon's study evaluated the repopulation of CD4+ T cells in the rectal mucosa of 6 sooty mangabeys after initiation of therapy with 2 antiretrovirals (ARVs), tenofovir and FTC. Half of the animals placed on therapy showed an increase in the percentage of these cells at this mucosal surface, suggesting that the immune mechanisms designed to reverse this damage are still active in non-pathogenic SIV infection.

Several groups also presented on the effect of HAART in human studies. Jason Baker from Haase's laboratory looked at both the pre-treatment levels and the increase in CD4+ T cells in the lamina propria of 8 people with acute HIV infection, 15 who were pre-symptomatic, and 7 who had AIDS (Abstract 41). In 12 of these individuals that were followed for 6 months, he observed only a 10% increase in the CD4+ T cells in the gut after starting HAART. These same individuals had an average 32% increase in CD4+ T cells measured in peripheral blood samples.

Another study looked at five individuals on HAART for at least three years and found that even after this length of time there was not a significant replenishment of CD4+ T cells in the gut. This second study, presented by Jason Brenchley from the VRC, also found in HIV-infected individuals on HAART, 10 times more HIV-infected CD4+ T cells in the gut than in peripheral blood (Abstract 38).

This finding suggests to Brenchley that the local concentration of ARVs in the gut may be much lower than in blood, allowing viral replication at mucosal surfaces to continue at high levels. This residual viral replication along with the continued depletion of the CD4+ T cells at mucosal surfaces can hinder immune reconstitution. Brenchley proposed that the varying concentration of ARVs could be due to the high number of cells in the gut that express P-glycoprotein, a toxin pump found on both lymphocytes and epithelial cells known to expel protease inhibitors.

"This is a possible explanation for why the CD4+ T cells aren't restored," he says. Brenchley is now looking at the level of viral replication in the GI tract in more HIV-infected individuals on HAART and is also comparing ARV concentrations to further determine if this is what impairs reconstitution of the CD4+ T cells.

There was also a study at CROI from Satya Dandekar at the University of California Davis on HIV-infected individuals who are able to restore CD4+ T cells in the gut (Abstract 39). Dandekar looked at 50 HIV-infected individuals that initiated HAART either during early infection (defined as within 4-6 weeks after exposure) or during chronic infection when CD4+ T cell counts dip below 400 cells/ml of blood. The 30 individuals who started HAART during early infection had significantly greater increases in CD4+ T cells during the first 2 years of therapy as compared to those who didn't begin HAART until later. One person was actually able to restore cells to pre-infection levels.

"Studies suggest that treatment during acute infection is better because it may attenuate the degree of depletion that would have otherwise occurred," says Brenchley. "It would probably work best if you could start as early as possible. This could be why post-exposure prophylaxis works so well."

And since the recovery in the gut differs significantly from that in the blood, Dandekar suggests HIV-infected individuals should be monitored by mucosal responses to obtain a better measure of how well HAART is working.

In another presentation, Binhua Ling of the Tulane National Primate Research Center reported on the restoration of CD4+ T cells in a colony of Chinese rhesus macaques, a sub-species of macaque that is genetically different from the Indian macaques most commonly used in these studies (Abstract 40). Ling began using this sub-species because of a lack of available Indian macaques but soon noticed that it took many of these animals 2-3 times longer to develop AIDS. Ling followed 10 SIVmac239-infected Chinese macaques, 4 of which were classified as long-term non-progressors (LTNPs) during the acute and chronic stages of infection. She found little difference between the two groups during the acute phase but after about six months, during chronic infection, the LTNPs had a greater ability to restore and maintain CD4+ T cells in the GALT.

Further experiments by Ling showed that depletion of the CD8+ T cells in two of these LTNP animals prevented a repopulation of the CD4+ T cells. Preston Marx, a virologist at Tulane who worked on this study suggests that immune control, mediated by the CD8+ T cells in the mucosal tissues, is required to suppress the virus for CD4+ T cells to be replenished. He and others are now looking to see if there is a specific MHC allele associated with suppression of such a highly pathogenic virus in these LTNP Chinese macaques.

Marx believes this model will be useful in understanding what allows some HIV-infected individuals to have markedly slower disease progression. "No one knows what the immunopathology is for someone who will be a long-term non-progressor," he says. "We hope that by looking more closely at these animals we can learn something that helps understand this process in humans."

Pep for PrEP

Using ARVs to prevent HIV transmission, a concept known as pre-exposure prophylaxis (PrEP), has attracted much attention over the past year. Much of this had to do with the closure of two trials testing the efficacy of the ARV tenofovir in blocking HIV transmission. The premise for these trials is simple; people at high risk of HIV infection swallow a single pill of tenofovir each day to see if it lowers their rate of acquiring HIV when accompanying other risk-reduction behaviors. The biological plausibility of this approach to HIV prevention was demonstrated with tenofovir in monkeys several years ago by the drug's manufacturer Gilead Sciences, but clinical trials testing PrEP only recently began in several countries. Although trials are still ongoing in the US, Thailand, Botswana, Peru, and Ghana, the trials in Cambodia and Cameroon were stopped last year after activist pressure.

At CROI, those following PrEP research received some promising news. Researchers from the US Centers for Disease Control and Prevention (CDC) showed that a combination of two ARVs, an idea now being called combo-PREP, may be even better at blocking HIV transmission in rhesus macaques (Abstract 32LB). Walid Heneine of the CDC presented data on the efficacy of tenofovir and FTC, two non-nucleoside reverse transcriptase inhibitors, in preventing rectal SHIV (a hybrid of SIV and HIV) transmission in six macaques. The animals received a subcutaneous injection of 22 mg/kg of tenofovir and an oral dose of 20 mg/kg of FTC for 9 days prior to viral challenge.

The SHIV challenge virus was administered repeatedly at a dose equivalent to 3.8x105 virus particles using the repeated rectal challenge model. All 9 control animals were infected after at least 4 exposures, while all treated animals remained free of infection after a total of 24 rectal inoculations. Three months after stopping tenofovir/FTC treatment, all these animals still remain uninfected.

A presentation at last year's CROI showed that tenofovir was able to delay but not entirely prevent infection in macaques when challenged with SHIV using this same model. So the results from combo-PrEP were particularly exciting. A possible explanation for the notably improved response to the tenofovir/FTC combination is the exposure of the drug at mucosal surfaces. Myron Cohen of the University of North Carolina presented data from his laboratory showing that the concentration of FTC is 600 times higher in the human female genital tract than in blood, making it a promising choice for PrEP. In order to determine how much this additional drug contributes to the efficacy of the combo-PrEP regimen, Heneine and colleagues also tested the efficacy of FTC alone in six macaques. They found that after 10 viral challenges 4 of the monkeys remained uninfected.

These impressive results suggest that this combination could be highly effective in preventing sexual transmission of HIV. Both researchers and prevention activists were enthusiastic about this study and some groups have already called for clinical trials with this regimen. Gilead has already licensed the co-formulated version of these two ARVs in a single pill known as Truvada and this makes it even more appealing for human studies. Although efficacy in this repeated rectal challenge model does not guarantee similar results in clinical trials, "it seems it may more closely approximate what is happening in humans," says Susan Buchbinder of the San Francisco Department of Public Health, who applauded the research.

"It is likely in my mind that antiretroviral therapy will serve as part of our overall HIV prevention strategy," said Cohen in his presentation on the future of HIV prevention.

Blocking superinfection

Ever since the first reports of HIV superinfection, including the announcement by Bruce Walker of Harvard Medical School at the 2002 International AIDS Conference in Barcelona, vaccine researchers have been on notice. The idea that the antibodies produced in response to HIV infection are not capable of neutralizing a virus that differs only slightly was great cause for concern. But based on research presented at CROI there is still hope that such protection does occur after a certain length of infection.

Evidence from both macaque studies and documented cases in humans indicate that infection with a second virus is most likely to occur within a window period of primary infection. There aren't any reported cases of superinfection in cohorts of chronically-infected patients, according to Robert Grant of the Gladstone Institute in California. Grant presented data on the neutralizing antibody responses to various viruses in a cohort of HIV-infected individuals and their HIV-infected partners (Abstract 92). All partners in this study had genetically distinguishable viruses at baseline, were not taking ARV therapy, and had established their partnership after becoming infected. The partners engaged in an average of 200 unprotected episodes of intercourse during this study.

Of the 9 individuals that were recently infected there were 5 cases of superinfection, while among the 12 individuals that were HIV infected for more than 2 years, there was no evidence of superinfection.

Grant found that when he measured the neutralizing antibody responses in the superinfected individuals, they had higher titers to their partner's virus than to autologous virus but that overall the responses were weak and narrow. In the individuals that were exposed but remained infected with only a single virus, the neutralizing antibody responses were cross-protective to autologous virus, virus from their partner, as well as to 10 epidemiologically unrelated viruses from other individuals. This finding led Grant to conclude that neutralization of the partner's virus might be blocking systemic superinfection in chronically HIV-infected individuals. Grant and his colleagues are now analyzing genital mucosal sites in these individuals to see if the antibodies created at these surfaces are blocking infection or if a superinfection is localized within the genital compartment.

The results of this study were good news to researchers but emphasized again the importance of finding a vaccine that can induce broadly neutralizing antibodies.

Ways around pre-existing immunity

Lawrence Corey of the HIV Vaccine Trials Network was given the task of summarizing the status of AIDS vaccine research at this year's CROI. He highlighted in his discussion the eight viral vectors that are currently being evaluated in clinical trials. Of these the one that is furthest along in clinical testing and perhaps holds the greatest promise, is the adenovirus serotype 5 vector (Ad5). But a major obstacle for this vector is the prevalence of pre-existing immunity (PEI; see Immunity's yin and yang).

"Apart from the problem with pre-existing immunity, Ad5 appears to be an excellent vaccine vector," said Dan Barouch of Beth Israel Deaconess Medical Center in Boston at CROI. One strategy for getting around issues with pre-existing immunity is to use a naturally-circulating adenovirus that is not as prevalent worldwide, such as Ad35. However in preclinical studies this vector is much less immunogenic than Ad5, according to Baruch.

To circumvent problems with PEI and maintain the immunogenicity of the Ad5 vector, Barouch and his colleagues have been tinkering with the hexon-capsid protein of Ad5, which is the primary target for neutralizing antibodies against the virus (Abstract 179LB). This group conducted experiments where they replaced the hypervariable regions of the hexon protein with genes from the corresponding region in the Ad48 virus and compared the responses to the unadulterated Ad5 vector.

In mice with high levels of PEI to Ad5 the cellular immune responses, as measured by tetramer binding assays, to a dose-limiting immunization with the traditional Ad5 vector expressing SIV Gag were blunted significantly. But when the chimeric vector was tested, the mice with high PEI had responses similar to those seen in mice without any PEI. Even when the chimeric vector was administered as both a prime and boost in mice with high PEI, both immunizations were highly immunogenic.

Barouch in partnership with researchers at Crucell, a biotechnology company in the Netherlands, then administered a 1011 viral particle dose of both the hexon-chimeric vector and the Ad5 vector expressing SIV Gag in rhesus macaques with and without anti-Ad5 immunity. The Gag-specific cellular immune responses (quantified by IFN-g  ELISPOT assays) elicited by the traditional Ad5 vector were suppressed by more than 4-fold in animals with high PEI but remained unchanged in those who received the hexon-chimeric vector.

These manipulations of the Ad5 vector look like a promising route for avoiding problems with PEI in populations where it is prevalent, and Barouch suggests that this approach is ready to move into clinical trials soon. He did, however, also allude to potential large-scale manufacturing issues with the chimeric vector.