Everything from Cause to Cure

Research presented at the biannual International AIDS Society Conference ran the gamut from studying early HIV infection to the search for a cure

By Kristen Jill Kresge

With the theme “From Cause to Cure,” the 5th International AIDS Society (IAS) Conference on HIV Pathogenesis, Treatment and Prevention, which was held from July 19-22 in Cape Town, South Africa, brought together more than 7,500 delegates to discuss a range of questions regarding everything from the earliest events of HIV infection to viral eradication. While the meeting covered the soup to nuts of HIV infection, Françoise Barré-Sinoussi, co-recipient of the Nobel Prize for the discovery of HIV, cited two main challenges plaguing researchers in her talk at the opening ceremony. “One challenge we have is to develop a vaccine, another is to have a cure for AIDS,” she said.

By no means are these small challenges. Research on viral eradication seems to have picked up steam recently, due in part to new funding from the US National Institute of Allergy and Infectious Diseases (NIAID), but it remains a formidable task. Barré-Sinoussi and the IAS continue to shed light on this effort. Next year, Barré-Sinoussi is organizing a two-day workshop in advance of the biannual International AIDS Conference in Vienna called “Towards a cure, HIV reservoirs and strategies to control them.”

In vaccine research, some of the emphasis has shifted to understanding the basic immunology of HIV and using this to design improved vaccine candidates. Efforts continue to be focused on understanding the earliest events of HIV infection, as well as identifying properties that allow some individuals to control HIV immunologically. “This is a virus that can actually be controlled by the immune system,” said Bruce Walker, director of the Ragon Institute, in his plenary talk at the conference, adding that “a vaccine for HIV is possible, but still a long way off.” Anthony Fauci, director of NIAID, said that although the field is moving more toward basic research, “It’s not going to slow things down. I think it’s [actually] going to speed things up.”

Meanwhile, researchers continue to focus on implementing existing HIV prevention strategies, such as adult male circumcision. There is also a growing chorus of support for sustaining and increasing the availability of antiretroviral therapy (ART) and initiating treatment earlier in the course of HIV infection, both to save lives and prevent new infections from occurring.

In the midst of a global recession that threatens the sustainability of HIV/AIDS funding, the need to maximize the benefit of existing HIV treatment and prevention funding was a pervasive message at the conference. Stefano Bertozzi, who recently joined the Bill & Melinda Gates Foundation as HIV director, suggested that the economic crisis might, paradoxically, help improve program efficiency and he called for better management and integration of services. “For the last 25 years we’ve had an emergency response to this epidemic,” said Bertozzi. “We have to switch that thinking.”

When to start

The approach to treating HIV has changed dramatically over the past 25 years. There are now more than 30 licensed antiretrovirals (ARVs), and combination regimens of these drugs. Effective therapy can reduce viral load to below detectable levels by standard assays and often even completely halt ongoing viral replication. Fauci said that now a newly HIV-infected 20-year-old who receives appropriate treatment has a life expectancy of at least 69 years. “The results are striking, historic, and to some degree unprecedented,” he said.

Over time, therapeutic strategies have also changed, with researchers see-sawing between early and delayed initiation of ART. Early on, the approach was to hit hard as early as possible to try to stop the virus from wrecking the immune system. However, clinicians eventually became concerned about the toxicity of ARVs and tended to delay initiation of therapy until a person’s health began to decline. Also, because the availability of ARVs was severely limited in developing countries, guidelines were devised so that therapy was not administered until a person’s CD4+ T-cell count dropped below 200 in a microliter of blood (the clinical definition of AIDS).

Now, evidence is accumulating that suggests starting therapy earlier in the course of HIV infection is beneficial. Wafaa El-Sadr, a professor in the department of epidemiology at Columbia University, presented data from the SMART study, which was conducted at more than 300 clinical research centers in 33 countries and compared the clinical outcomes of nearly 5,500 HIV-infected individuals who were randomized to either start ART early in the course of their infections, or later, when their CD4+ T-cell levels fell below 250. Rates of serious AIDS- and non-AIDS-related events were 2.4 per 100 person years among those who started therapy early, versus 4.4 for those who delayed treatment.

Earlier initiation of therapy may be beneficial because of its ability to suppress ongoing rounds of viral replication that lead to chronic activation of the immune system, which can be detrimental over the long term. Researchers previously thought that during the eight to 10-year period following infection, the virus was clinically latent, but more recent evidence suggests clinical latency is a “misconception,” according to El-Sadr. Ongoing viral replication during the period of asymptomatic HIV infection induces “inflammatory changes that are associated with an increased risk of mortality,” she said. Inflammation is an important cause of organ damage, disease progression, and death, added El-Sadr. “HIV is much more toxic than any drug you can throw at it,” said Julio Montaner, president of IAS.

Another study provided additional evidence that starting therapy earlier can improve clinical outcomes. The study, known as CIPRA HT 001, involved 816 HIV-infected adults in Haiti with CD4+ T-cell levels between 200 and 350. Half of the participants were randomly selected to start treatment within two weeks of enrollment, while the remaining volunteers did not receive ART until their CD4+ T-cell counts dropped below 200, in accordance with the current treatment guidelines set by the World Health Organization (WHO). At an interim review, the study’s data safety monitoring board (DSMB) found that starting treatment at CD4 counts between 200-350 improved survival rates compared with deferred treatment. The mortality rate was four times higher among volunteers who deferred treatment until their CD4 count dropped below 200. Twice as many people in the group that received deferred therapy also developed tuberculosis (TB) during the study. Based on these findings, the DSMB recommended the trial be stopped and that all volunteers be offered ART.

Additional data indicates that ART may be an effective strategy in controlling TB. Keren Middelkoop, a principal investigator at the Desmond Tutu HIV Center, presented data from a follow-up HIV/TB survey conducted in a township community in Cape Town, South Africa, in 2008. The first survey of this community was conducted in 2005 and included 762 participants. At that time approximately 12% of those HIV infected were receiving ART. In 2008, after the South African government instituted a large-scale ARV program, a second survey of 1,251 people was conducted and 31% of those HIV infected who qualified for treatment were receiving it.

Middelkoop reported that HIV prevalence rose from 23% in 2005 to 25% in 2008. But after adjusting for age, sex, and HIV infection status, the population-level TB prevalence in 2008 was significantly lower—it dropped from 3% in 2005 to 1.8% in 2008. This decline was driven primarily by a reduction in prevalence of previously undiagnosed TB in HIV-infected individuals from 9.2% to 3.6%. During this time there were no changes in any of the TB programs or immigration levels that could have led to the decline in prevalence. Middlekoop concluded that the strong temporal association between reduction in TB prevalence and the initiation of an ART program suggests it is possible that ART reduces TB transmission.

“Everything seems to point toward earlier therapy,” said Fauci. This suggests that perhaps the current treatment guidelines may need to be revised. The IAS has already altered their recommendation, suggesting therapy be commenced at <350 CD4+ T cells, and many think the WHO may soon modify their guidelines as well. Most developing countries follow the WHO’s recommendation, so if they adopt earlier treatment, many more people would qualify for ART, dramatically increasing global treatment costs. But Fauci said earlier therapy is still a cost-effective strategy. “It doesn’t actually cost more money [because] it’s twice as expensive to care for people who don’t start early,” he said.

However, the optimism about early initiation of treatment was tempered by warnings about potential drug shortages in some countries that could jeopardize access to therapy, even for individuals already on treatment. Bertozzi said that there is still so much to be done to get treatment to people who will die without it that, “It’s really hard to imagine how [earlier treatment] could be a reality in the near future.”

Treatment as prevention

As researchers reconsider the optimal time to begin therapy, there is also an ongoing push to expand availability of ARVs to help stem the spread of HIV (see Test and Treat on Trial). “Highly active antiretroviral therapy is an essential tool to curb the growth of the pandemic,” said Julio Montaner, president of IAS and a pioneer of early initiation of treatment, who called therapy a “cost-averting intervention even in a fiscally challenging environment.” Bernard Hirschel, head of the HIV/AIDS unit at the University Hospital of Geneva, even declared that, “Treatment as prevention is the topic of the year.” He pointed to HIV transmission data from a recently published study of circumcision as a reason to explore ARVs for prevention (1). In this study, despite free condoms and HIV counseling, the likes of which he said would never be achieved outside of the setting of a clinical study, the HIV transmission rates in both the control and intervention groups were above 10%. “Is treatment the answer?” Hirschel asked. “I don’t know, but we better find out.”

In a study presented by lead author Patrick Sullivan of Emory University, researchers looked at the rates of HIV transmission in 2,993 serodiscordant couples from Rwanda and Zambia over almost a year and a half. During this time, 172 linked infections between couples occurred when the infected partner was not on ART, while only four occurred if the infected partner was on treatment (initiated at a CD4+ T-cell count less than 200). Based on this data, there was a 79% reduction in HIV transmission associated with the infected partner taking ART.

All four of the HIV transmissions that occurred when the infected partner was taking ART were from women to men, but Sullivan concluded that there were too few transmissions to reach a conclusion on whether relative rates of HIV transmission among serodiscordant couples on ART differ by sex. Sullivan also noted several caveats to this study, including the fact that viral load measurements were not routinely taken so they could not link transmission to viral suppression. He said that although these data can’t be generalized outside of these specific cohorts, they do “support the idea that treatment of patients with clinical indication may reduce HIV transmission.”

Others were more circumspect about the possibility of treatment as prevention. Sarah Fidler, senior lecturer at Imperial College in London, said directing treatment to the most infectious individuals would have the greatest effect on reducing transmission rates, but this requires identifying infected individuals within a few months after seroconversion. She noted that three main factors determine infectiousness: the viral load of the infected individual, whether or not the individual has other sexually transmitted infections (STIs), and the stage of their HIV disease. Fidler said that it is during primary or acute infection that an HIV-infected individual is the most infectious. A study involving serodiscordant couples in Rakai, Uganda, found that during primary infection, an individual is estimated to be 26 times more infectious than during the long duration of asymptomatic infection that follows (2). Infectiousness increases again during the late stage of infection (10-19 months before death), but to a lesser degree than during the acute phase. During acute infection, HIV-infected people are often not aware that they are infected, are likely to also have other STIs, and have “extremely high” plasma and genital tract viral loads, according to Fidler. Also, recently transmitted HIV variants are often biologically more transmissible.

The period of “high infectiousness” during primary infection was estimated in the Rakai study to last approximately three months after initial seroconversion. Fidler said that if the “high viral load is short-lived, the increased risk of transmission is also short-lived.” As no treatment guidelines call for initiation of therapy during acute infection, Fidler casted doubt on whether earlier treatment could substantially lower HIV transmission rates on a population level.

Early control of HIV

There is also an ongoing effort to better understand the earliest events of HIV infection. “We know an awful lot about the pathogenesis of HIV but there are still questions we can’t answer,” said Fauci.

Some researchers are focusing on the early stage of HIV infection in controllers—individuals who maintain viral loads of less than 2,000 HIV RNA copies/ml. HIV controllers are the subject of much research, which is of particular interest to vaccine researchers, but little is known about how these individuals control virus during acute infection. Toshiyuki Miura of the Institute of Medical Science at the University of Tokyo presented an analysis of host genetics and virus characteristics in a group of 18 controllers during acute HIV infection, so-called acute controllers. Based on human leukocyte antigen (HLA) typing, Miura reported that host genetics were not responsible for effective control of HIV—the frequency of acute controllers expressing any of the HLA class I alleles that are considered protective against HIV (B*13, B*27, B*51, and B*57) was actually significantly lower than among HIV controllers during chronic infection. Only two acute controllers expressed the B*57 allele.

Miura and colleagues then constructed chimeric viruses derived from the Gag-Protease proteins isolated from virus in acute controllers. Researchers observed that these chimeric viruses had reduced replicative capacity, attributed in part to cytotoxic T lymphocyte (CTL) escape mutations. Viral sequencing indicated that viruses in acute controllers had a higher frequency of drug resistant mutations than the viruses in progressors during acute infection. Two of the acute controllers who possessed the B*57 allele had a T242N viral escape mutation that affects viral fitness. Miura concluded from this small study that individuals who control viremia following acute infection are infected with viruses that have reduced replicative capacity that is attributable to the transmission of drug resistant strains or fitness costs that are the result of CTL escape mutations.

Host genetics and immune control

Walker and colleagues are one of the main groups studying HIV controllers to mine for clues that may inform vaccine development. Their research is also uncovering important characteristics about the role of host genetics in immune control of HIV. According to Walker, HLA B alleles are “really important in this disease,” as they greatly influence viral load. HLA class I alleles display viral peptides on the surface of infected cells that are recognized by T-cell receptors. There are 817 HLA class I B alleles, some of which have previously been associated with control of HIV. In a study conducted in KwaZulu-Natal, South Africa, Walker and colleagues did high resolution HLA typing of approximately 700 individuals. They found that some alleles, such as B*57, B*8101, and B*5801, are associated with protection against HIV and result in lower viral loads, while others such as B*5802 are associated with disease progression and higher viral loads. HLA distribution within the population largely reflects ethnic ancestry—both B*5801 and B*5802 are common in African populations. In this study, individuals with the B*5801 allele had a median viral load of 14,650 HIV RNA copies/ml, while individuals possessing a B*5802 allele had a median viral load of 75,200. Individuals with neither of these alleles had a median viral load of 33,000 copies/ml.

In another study, Walker and colleagues confirmed the observation that broader Gag-specific CD8+ T-cell responses are associated with lower levels of viremia, while conversely, broader Env-specific CD8+ T-cell responses were associated with higher viral load. The B*5801 allele primarily presents Gag epitopes, while the B*5802 allele predominantly presents HIV Env epitopes, explaining at least in part, why one allele is associated with protection and the other with progression.

Using a viral suppression assay, Walker and colleagues then compared the ability of CD8+ T cells freshly isolated from chronically HIV-infected individuals with high Gag-specific CD8+ T-cell responses, to those from individuals with similar viral loads yet low Gag-specific CD8+ T-cell responses. They found that the CD8+ T cells from high Gag responders were better able to neutralize HIV and inhibit viral replication. “There’s a qualitative difference depending on what part of the virus is being targeted,” said Walker. Further analysis showed that this is, in part, because Gag-specific CD8+ T-cell responses select for viral fitness mutations that cripple HIV.

To evaluate the population-level impact of epitope-specific immune pressure on HIV, Walker described efforts led by Philip Goulder, an immunologist at Oxford University, to conduct HLA typing and virus sequencing of Gag from 2,800 chronically HIV-infected individuals from nine cohorts, spanning five continents. Goulder and colleagues then analyzed the HLA prevalence and detection of escape mutations in HIV. This work, recently published in Nature, suggests that the more prevalent an HLA allele, the more immune escape viruses that are transmitted (3). “HIV is being shaped by the immune response [against it] on a global level,” said Walker, “and some protective epitopes are being lost.” He and the other study authors suggest that keeping up with the changing immunological landscape of HIV will be a challenge for vaccine researchers.

Host genetics can also impact HIV transmission. In research, which will soon be published in the Journal of Virology, Walker and colleagues found that among mothers who express protective HLA alleles, transmission of HIV to their infants is less common. Additionally, infants born to mothers with protective HLA alleles, or who themselves had protective alleles, were more likely to be slow progressors.

T-Cell Immunity and HLA Molecules.Human leukocyte antigen (HLA) class I alleles, which are determined by host genetics, display HIV peptides or epitopes that enable immune recognition and killing of HIV-infected cells by CD8+ T cells. There are 817 HLA class I B alleles, some of which have previously been associated with control of HIV.

The status of HIV prevention

Ronald Gray, professor in population and family planning at Johns Hopkins University, pointed out in a plenary talk on the state of HIV prevention that out of 29 completed trials evaluating the efficacy of different biomedical interventions to date, only four had shown significant success (three evaluating adult male circumcision and the other evaluating treatment of sexually transmitted infections to reduce HIV risk). Five showed possible harm. “HIV prevention trials are extremely difficult and expensive,” said Gray. As such, he challenged researchers to urgently improve the design of trials, to do a better job of pre-clinical screening and assessment, and to collect better pre-trial HIV incidence estimates.

Gray suggested that treating sexually transmitted infections such as herpes simplex virus (HSV)-2 is still a priority, but its effect on the HIV epidemic may be limited. However, Jairam Lingappa, medical director of the Partners in Prevention HSV/HIV study, presented data showing that treatment of HSV infection with acyclovir delays HIV disease progression by 17-19%. Lingappa and colleagues analyzed 3,381 of the HSV/HIV-infected partners in the study with median CD4+ T-cell counts of about 400. The primary analysis was conducted based on a composite endpoint of reaching a CD4+ T-cell count of less than 200, starting ART, or dying during the study. Of those receiving acyclovir, 284 individuals reached this composite endpoint compared to 324 individuals who received placebo—a statistically significant difference. Lingappa said there are multiple interpretations for this observation. He presumed that the effect is mediated by the observed 0.25 log reduction in HIV RNA copies/ml of plasma observed in the acyclovir group, though he did also refer to recent studies that have shown that acyclovir has direct antiretroviral activity in vitro. He concluded that this presents a new concept for delaying HIV disease progression without ART. However, a question was raised about a possible downside of this observation. If ART can reduce HIV transmission, delaying initiation with acyclovir could only prolong the period of time that an HIV-infected person is able to transmit HIV to others.

Another possible strategy for reducing HIV infection risk may be vaccination against human papilloma virus (HPV). Jennifer Smith, a researcher at the University of North Carolina in Chapel Hill, reported that men in a circumcision trial in Kenya who were HPV infected had an 80% higher risk of HIV infection through 42 months, even after controlling for circumcision status and HSV-2 infection. Smith said there were many possible explanations, including that HPV could potentially induce specific local cytokines, such as macrophage inflammatory protein-3 or interleukin-8, which may increase HIV susceptibility, and added that HPV prevention efforts may help reduce HIV acquisition. Two vaccines against HPV are now licensed for use in many countries throughout the world.

Circumcision Progress

Multiple sessions at the 5th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention focused on progress in implementing adult male circumcision (AMC) for HIV prevention. Kawango Agot, of the Universities of Nairobi, Illinois, and Manitoba Project, gave an overview of progress on this front in the Kisumu district of Kenya. Since the Kenya National Voluntary Male Circumcision Program was initiated in November 2008, around 30,000 males have been circumcised through the public system, largely with support from donor agencies. One obstacle to implementing circumcision programs was the lack of trained medical professionals. In June, Kenya revised its health regulations to allow nurses to conduct the surgical procedure, and Agot’s group has found that after training, adverse events following AMC did not differ if the procedure was performed by nurses.

It is still unknown whether AMC reduces rates of HIV infection in men who have sex with men (MSM), but Tim Lane of the University of California in San Francisco, presented data from a study in Soweto, South Africa, showing that uncircumcised MSM had nearly five times higher HIV infection rates than their circumcised counterparts. Lane said that about 40% of men in the Soweto cohort also had sexual relations with women, therefore, “reducing the risk of HIV among MSM could benefit entire communities.” He called for further research to assess the acceptability of an AMC trial for HIV prevention among MSM.

—Daisy Ouya, Program Manager, Information, Education and Communication for IAVI in Nairobi, Kenya


1.Lancet 374, 229, 2009
2. J. Infect. Dis. 198, 687, 2008
3. Nature458, 641, 2009